REVIEW ARTICLE

Drugs 44 (6): 946-962. 1992 00 12-6667/92/00 12-0946/$08 .50/0 © Adis International Limited. All rights reserved. OAUI221

Interferons in Multiple Sclerosis A Review of the Evidence Hillel S. Panitch Neurology and Research Services, VA Medical Center and Department of Neurology, University of Maryland School of Medicine, Baltimore, Maryland, USA

Contents 946 948 948 949 95/ 954

956

957

Summary

Summary 1. Development of the Interferons 2. Clinical Applications of Interferons in Conditions Other Than MS 3. Early Clinical Trials of Interferons in MS 4. Immunoregulation by the Interferons 5. Additional Evidence for a Pathogenic Role of Interferon Gamma in MS 6. Current Therapeutic Trials of Recombinant Interferon Beta 7. Conclusions

Multiple sclerosis (MS) is an inflammatory demyelinating disease of the central nervous system, characterised clinically by relapses and remissions, and leading eventually to chronic disability. Despite an enormous amount of research, the cause of MS remains unknown; however, pathological, genetic, and immunological features have been identified that suggest the disease has an autoimmune basis. Accordingly, current therapy of MS includes corticotrophin or corticosteroids for acute exacerbations, and more potent immunosuppressive drugs for severe cases unresponsive to steroids. All of these agents can cause serious adverse reactions. There is an urgent need for immunotherapy that is less toxic, that can be given early and perhaps indefinitely, and that will prevent relapses and progression of the disease. Our current knowledge of the effects of interferons (IFNs) in MS is based on the results of laboratory research and clinical therapeutic trials carried out over the past decade. Existing evidence points to the conclusion that the effects of the IFNs in MS are mediated by immunoregulatory rather than antiviral or nonspecific mechanisms. Administration of IFN'Y increases the exacerbation rate, and IFN'Y as well as other cytokines may be involved in the pathogenesis of MS lesions. In contrast, studies of IFNi3 show that it tends to inhibit the activity of IFN'Y and appears to prevent disease activity. Intrathecal administration of IFNi3, although effective, is cumbersome and potentially hazardous. A large multicentre placebo-controlled trial of systemic recombinant IFNi3 was recently conducted in the US, and the results of the first 2 years of treatment were considered sufficiently encouraging that an application for licensingwas submitted to the Food and Drug Administration in June 1992. If approved, it will be the first new agent licensed for clinical use in MS in over 20 years. The study will continue under double-blind conditions for at least another year, and a second trial of systemic recombinant IFNi3 therapy is also in progress. These studies should provide definitive answers to Questions about the role of IFNs in the pathogenesis of MS, as well as the place of recombinant IFNi3 as an effective therapeutic agent.

Interferons in Multiple Sclerosis

Multiple Sclerosis (MS) is a disease of young adults which is characterised pathologically by multiple areas of inflammation and demyelination in the brain and spinal cord, and clinically by recurrent attacks (exacerbations) of neurological dysfunction , often leading to progressive physical impairment. The cause of MS is unknown. Epidemiological studies suggest the involvement of a transmissible agent, but no such agent has ever been reproducibly demonstrated in MS patients . Recent immunological studies have provided evidence for immunoregulatory abnormalities in MS, and have been widely interpreted as suggesting that MS is mediated by an autoimmune process (Dhib-Jalbut & McFarlin 1989; McFarlin & McFarland 1982; Ransohoff 1992). As in most diseases of presumed autoimmune aetiology, it is not known how immunity to self antigens is initiated, although viral infection of a genetically susceptible host during childhood or adolescence may be responsible. Certain features of MS specific to the central nervous system (CNS), are the pathological hallmarks of inflammation and demyelination in the white matter, increased intrathecal synthesis of immunoglobulins, the presence of oligoclonal immunoglobulin G (IgG) in the cerebrospinal fluid (CSF), and lack of pathological changes in other organ systems. Studies of peripheral blood lymphocytes, however, suggest that the immunoregulatory defect is a systemic one, while the antigenic target is confined to the CNS (Hafler & Weiner 1989). For example, activated T cells that specifically recognise myelin basic protein (MBP) are present in peripheral blood (Allegretta et al. 1990; Hafler et al. 1985), suppressor cell function is defective in patients with active disease (Antel et al. 1978, 1986), and cytokine production by peripheral blood cells may be abnormal (Beck et al. 1988; Trotter et al. 1988). The therapeutic effect of corticosteroids and other immunosuppressive agents given systemically indicate a primary action on peripheral immune responses. However, the striking occurrence of blood-brain barrier defects in active MS, as demonstrated by gadolinium-enhanced magnetic resonance imaging (MRI) [Gonzalez-Scarano et al. 1987; Harris et al. 1991; Ker-

947

mode et al. 1990], suggests that the effects of systemically administered agents need not be confined to the peripheral immune system. In recent years, research interest has centred around the interaction of T cell receptors, major histocompatibility complex (MHC) molecules on antigen-presenting cells, and one or more myelin antigens as the basis of the autoimmune process (Waksman 1985). The use of T cell ant igen receptor subtypes is reported to be limited (Kotzin et al. 1991 ; Oksenberg et al. 1990), and the immune response is further restricted by particular MHC determinants , for example HLA-DR2 and HLA-DQwl in patients of Northern European origin (McFarlin 1988; Vartdal 1989). The autoantigen involved in the MHC receptor-antigen complex is unknown , but MBP has received the most intense scrutiny because of its central role in experimental allergic encephalomyelitis (EAE), the best animal model of MS. Myelin proteolipid protein, which can also induce EAE (Tuohy et al. 1988), is another candidate antigen. One investigative group has reported that immunodominant epitopes of MBP are recognised with relative specificity by the T cell receptor (Wucherpfennig et al. 1990). In contrast, others have found that T cell receptor usage is not restricted to a particular MBP epitope, and may vary among MS patients, or even within a single patient (Ben-Nun et al. 1991 ; Martin et al. 1990; Richert et al. 1991). It is also possible that the immune response in MS is less specific than previously thought, and may involve non-MHC-restricted T cells expressing the 'Yo-receptor (Freedman et al. 1991), or the recently discovered 'superantigens', which could bypass the requirement for a particular myelin antigen by activating T cells through nonspecific linkage of their receptors to MHC molecules (Rudge 1991). The picture is further complicated by a myriad of endogenous immunoregulatory processes in the systemic and CNS compartments, which can modulate antigen recognition, cellular activat ion, trafficking of activated cells, and expression of effector activity. These include various T cell subsets, antigen presenting cells, cytokines, and neuropeptides. What then is the place of the interferons (lFNs)

948

in this complex system, and how can their use in the treatment of MS be justified? Available evidence suggests that the IFNs are most likely to influence MS disease activity by virtue of their effects on regulation of MHC molecules and presentation of antigen to T cell receptors. In this review, we will summarise the background from which current knowledge of IFN function has developed, outline effects of the IFNs on immune processes thought to be relevant to MS, and present the results of recent and ongoing clinical therapeutic trials , which seem to point to systemically administered IFN,8 as a potentially useful therapeutic agent.

1. Development of the Interferons Natural IFN was described in 1957 (Isaacs & Lindenmann 1957) as a protein secreted by virusinfected cells which acts' on other cells to prevent them from becoming infected. Subsequently, two principal types of IFN were recognised: Type I (including leucocyte or alFN, and fibroblast or ,8IFN), and Type II (immune or -yIFN). The advent of recombinant DNA technology in the early 1980s permitted the cloning of IFN genes and their products, and precipitated a revolution in IFN research (Pestka 1983). The gene for human IFNa was among the first to be sequenced and cloned, and a wealth of information on IFN genes and the molecular structure of the IFNs and their receptors rapidly appeared. IFNa can be synthesised by many cell types, but principally by lymphocytes and macrophages exposed to viruses or viral proteins. At least 24 different gene sequences, 15 of which are naturally expressed, code for subtypes of human IFNa (Fleischmann et al. 1988). There are differences in biological activity among the subtypes, the significance of which is not yet well understood. Natural IFN,8 is produced by fibroblasts , and probably many other cells, stimulated with synthetic or viral oligonucleotides (De Maeyer & De Maeyer-Guignard 1988c). In humans it consists of a single molecular species that, unlike most of the IFNa subtypes, is a glycoprotein. IFNs a and,8 are 30 to 40% homologous in terms of nucleic acid and

Drugs 44 (6) 1992

amino acid sequences, their genes are both encoded on chromosome 9, and they react with a common cell surface receptor (De Maeyer & De Maeyer-Guignard I988b). IFN-y is produced by activated T lymphocytes. It is a unique glycoprotein that shares no significant sequence homology with IFNs a or ,8, is genetically localised to a different chromosome, and utilises a different cellular receptor. IFN-yis so different from the other IFNs that it is now believed its antiviral and antiproliferative properties may be fortuitous, and that its primary function is as an immunoregulatory cytokine (Billiau 1987). This concept will assume more importance below, in attempting to understand the effects of the different IFN s in MS, and in considering IFN s a and ,8 as potential therapeutic agents. As large quantities of pure recombinant IFNs were produced for clinical and laboratory testing, it became possible to design clinical trials of the IFNs on a rational basis, and to devise experiments to clarify mechanisms ofIFN activity in the settings of viral infection , cancer and autoimmune disease. The IFNs are now known to be part of a vast cytokine network (De Maeyer & De MaeyerGuignard I988a) that includes Iymphokines, monokines , growth factors and certain peptide hormones. The relationships of these substances to each other, and to the cells of the immune system and the nervous system, are Byzantine in their complexity. Experimental findings in vitro often fail to predict the behaviour of the IFNs in vivo, making controlled clinical trials essential (Johnson & Panitch 1990). In fact, progress in understanding the biological properties and immunoregulatory functions of the IFNs has evolved over the past decade largely as a product of their use in clinical trials.

2. Clinical Applications of Interferons in Conditions Other Than MS Use of natural IFNa to treat human disease became feasible with the development by Cantell and Hirvonen (1978) of large scale techniques for producing a partially purified active product. Although the purity of this product was less than 1%,

949

Interferons in Multiple Sclerosis

clinical trials were undertaken in viral, neoplastic and autoimmune diseases. Other preparations used in early studies included natural IFN~ from cultured fibroblasts, and Iymphoblastoid IFN, a product of lymphocytes transformed by Epstein-Barr virus. Clinical research proceeded slowly in the 1960s and 70s because of the limited quantities of IFN available, and was largely disappointing in terms of therapeutic success. However, such studies generated a great deal of information on the biological properties of the IFNs, as well as their pharmacokinetics and side effects, and laid the foundation for later trials of recombinant IFNs, which have begun to show promising results (Volz & Kirkpatrick 1992). Until recently, the IFNs were regarded as potential miracle drugs in search of a disease, but during the past decade therapeutic effects have been described in several conditions (Fleischmann et al. 1988). IFNa inhibits influenza and rhinoviruses, and is partially effective as prophylaxis against the common cold when given intranasally. IFNs a and ~ have been shown to modify the course of herpes keratitis and other herpetic infections, although they do not eliminate persistent infection. Recombinant IFNa has also been reported to improve recovery from chronic active hepatitis. Unfortunately, trials in patients with ~IDS have demonstrated little or no specific therapeutic effect against HIV infection (Krown 1986). The value of IFNs in tumour immunotherapy is essentially limited to virus-associated tumours such as hairy cell leukaemia, condylomata acuminata and AIDS-related Kaposi's sarcoma (Fleischmann et al. 1988). Recombinant IFNa has been approved by the Food and Drug Administration and is commercially available for these 3 conditions. It was also recently approved for treatment of hepatitis C, and encouraging preliminary results have been obtained in non-Hodgkin's lymphomas, malignant melanoma and juvenile laryngeal papillomatosis. In addition, natural IFNa may exert a beneficial effect in some cases of subacute sclerosing panencephalitis when given intrathecally (Panitch et al. 1986; Smith et al. 1986). IFNoy is effective in chronic granulomatous disease and was

recently approved for this indication. It has also undergone trials in toxoplasmosis, rheumatoid arthritis and other systemic autoimmune diseases. IFN~ has thus far not been approved for any medical indication. Clinical testing of the IFNs elicited a pattern of adverse effects which is quite stereotyped and independent of the type of IFN used . At first these were thought to be caused by contaminants of the natural preparations, but because the same symptoms occur with recombinant IFNs, they are now recognised as unwanted effects of the IFNs themselves . The most common adverse events are 'flu-like symptoms of fever , chills, nausea, anorexia, fatigue, malaise, myalgia and headache (Rohatiner & Farkkila 1988). Bone marrow suppression and neurological side effects have occurred with high doses, usually exceeding 100 million Uzday, administered to cancer patients. Although the neurological effects are rare , they can be severe. Weakness, fatigue, depression, cognitive deficits and encephalopathies associated with slowing of EEG activity have been reported (Meyers et al. 1991; Rohatiner & Farkkila 1988; Rohatiner et al. 1983; Suter et al. 1984). The mechanism of these neurological complications is puzzling because the IFNs are thought to be relatively excluded from the CNS by the blood-brain barrier (Habif et al. 1975; Smith et al. 1985). Fortunately, adverse effects are dose-dependent and reversible with cessation of therapy. However, fevers as high as 39 to 40·C assume potential importance in MS patients, whose signs and symptoms may be temperature-sensitive. Furthermore, neurotoxic effects may interfere with evaluation of clinical efficacy in trials of IFNs in MS and other neurological diseases.

3. Early Clinical Trials of Interferons in MS The rationale for treatment of MS with IFNs was initially based on 2 lines of reasoning: firstly, that recurrent or continued disease activity might be related to viral persistence or latency in the CNS (Booss & Kim 1990; Cook & Dowling 1980); and

Drugs 44 (6) 1992

950

secondly that findings of defective IFN synthesis and natural killer cell activity had pathogenic significance (Merrill et al. 1982;Neighbour et al. 1982) and could be reversed by treatment with exogenous IFN. In 1981, Jacobs et al. reported an open unblinded trial of natural IFNfj in 20 patients with MS randomised to receive either intrathecal IFN or no treatment. A dose of 1 million U was given twice weekly for 4 weeks, then monthly for 5 months, after which the patients were observed for an additional 18 months. The frequency of attacks was lower in the IFN-treated group, and side effects were reportedly minimal despite the occurrence of CSF pleocytosis in many of the patients (table I). These results were sufficiently encouraging to support the design of a larger and better controlled double-blind study in which 69 patients with clinically definite relapsing-remitting MS were randomised to receive either 1 million U of natural IFNfj by lumbar puncture, or sham lumbar puncture, 9 times over 6 months (Jacobs et al. 1986,

1987). The patients were then followed for 2 years. Randomisation was effective, blinding was well maintained by the use of indomethacin to mask adverse effects, and complications were minimal. The patients treated with IFNfj had a significantly lower exacerbation rate during the study than the placebo group (0.76 vs 1.48attacks/year, p < 0.001), although there was no difference in progression between the groups. Despite the apparent success of this trial, intrathecal IFN has not become an accepted mode of therapy because of the need for multiple lumbar punctures and the possibility of inducing infection, adhesive arachnoiditis, or other complications associated with introducing a foreign protein into the subarachnoid space. Furthermore, in contrast to these studies, a more recent small but well-controlled trial (Milanese et al. 1990) reproduced the protocol of Jacobs et al. with a slightly different preparation of IFNfj. Relapse and progression rates increased during IFN treatment, casting some doubt on the value of intrathecal therapy. There have been 4 major studies of IFNa in

Table I. Principal clinical trials of natural interferons in MS Reference

Design

Interferon

Dosage (duration [months!)

Route

No. of patients

Clinical type

Results

AUSTIMS (1989)

db, pc

Natural a

SC

225

RR,CP

No effect of IFN

Bever et at. (1986)

open

Poly-ICLC induced

IV

18

CP

Jacobs et al. (1981)

db, pc

Natural {J

IT

69

RR

Improvement or stabilisation in 12 patients Fewer attacks in IFN group

Jacobs et al. (1981)

sb

Natural {J

IT

20

RR, RP, CS

Fewer attacks in IFN group

Kastrukoff et al. (1990) Knobler et al. (1984)

db, pc

Lymphoblasto id

SC

101

CP

db, pc, co

Natural a

3MU twice per week, then once per week [12] Variable; once per week to once per month [18] 1MU once per week, then once per month [6] 1MU twice per week, then once per month [6] 5 MU/day [6] 5 MU/day [6]

SC

24

No effect of IFN Fewer attacks in RR group

RR,RP

Abbreviations: IFN = interferon; MU = million units; IT = intrathecal ; IV = intravenous; SC = subcutaneous ; RR = relapsing-rem itting; RP = relapsing-progressive ; CP = chronic progressive; CS = chronic stable; sb = single-blind; db = double-blind; pc = placebocontrolled ; co = crossover.

Interferons in Multiple Sclerosis

MS, all involving systemic administration of the agent. In 1980, a double-blind, placebo-controlled trial of natural, purified IFNa was begun in 24 patients with clinically proven MS (Knobler et al. 1984). Because of the scarcity and high cost of the IFN, a crossover design was devised with each patient receiving either placebo or 5 million U of IFN by subcutaneous injection daily for 6 months, followed by a washout period, then by the alternate treatment and a second washout period. Although there were fewer exacerbations during the IFN phases of treatment, the results were not statistically significant. In a subgroup of 15 patients with strictly relapsing-remitting disease, there was a more pronounced trend favouring IFN (0.34 vs 1.14 attacks/year, p = 0.08), yet most of this apparent effect depended on changes in the relapse rates of 2 patients who were particularly responsive. Several IFN-mediated effects were found, including increased IgG synthesis in the peripheral blood and CSF, circulating immune complexes, and natural killer cell activation (Panitch et al. 1985; Rice et al. 1985), but none of these correlated with clinical efficacy. The AUSTIMS study (Australian Study of Transfer Factor and Interferon in MS) included 182 patients with either relapsing of progressive MS, randomised to receive either 3 million U of natural IFNa, transfer factor, or placebo by subcutaneous injection twice a week, then progressively less frequently over the 2-year study period (AUSTIMS Research Group 1989). The results were unambiguously negative, with no significant reduction in relapse rates or differences in disability scores in any of the groups. Lymphoblastoid IFN, consisting ofa mixture of IFNa subtypes and a small amount of IFN,8, was tested in 100 patients with chronic progressive MS (Kastrukoff et al. 1990). They were randomised double-blind to receive either 5 million U of IFN or placebo by subcutaneous injection daily for 6 months, and were followed for 2 years. There was no therapeutic effect in terms of neurological outcome or quantitative MRI scans . CNS IgG synthesis and the proportion of circulating CD5+ and CD4+ T cells increased (Kastrukoff et al. 1991),

951

whereas testing in vitro paradoxically revealed a striking decrease in pokeweed mitogen-stimulated IgG production in IFN-treated patients (O'Gorman et al. 1987). However, there was no relationship of laboratory findings to clinical outcome. In the first study of a recombinant IFN in MS, 98 patients received either 2 million U of IFNa-2 or placebo by subcutaneous injection 3 times per week for I year (Camenga et al. 1986) [table II]. On this low dose, neither therapeutic nor toxic effects were found , but a striking placebo effect occurred. Relapse rates fell dramatically in both the IFN and placebo groups. Natural killer cell activity increased during treatment not only in the patients who received IFN , but in the placebo-treated patients as well, a curious finding which remains unexplained (Hirsch et al. 1988). In summary, these 4 clinical trials of natural, lymphoblastoid, and recombinant IFNa yielded negative or equivocal clinical results, although they showed that the IFNs could be administered safely to patients with relatively mild MS. A somewhat different approach was taken by Bever et al. (1986) who treated a group of patients with chronic progressive MS with the interferoninducer poly-ICLC. This was an open study with no placebo-treated control group. The IFN induced by poly-ICLC includes a mixture of a, ,8 and 'Y subtypes, with IFNa predominating. Fever and other adverse effects were pronounced, and several patients experienced transient worsening of signs and symptoms after each intravenous infusion. Twelve of the 18 patients improved or stabilised during treatment, although the improvement was not maintained after the drug was stopped (table I). The investigators concluded that this form of therapy may be effective in selected patients, but that it is probably too toxic to be considered for long term general clinical use. Moreover, it has no apparent advantages over recombinant IFNa or IFN,8 which are much better tolerated.

4. Immunoregulatlon by the Interferons The studies described above were designed, and in some cases completed, before the immunoregulatory properties of the IFNs were fully appre-

952

Drugs 44 (6) 1992

Table II. Principal clinical trials of recomb inant interferons in MS

Reference

Design

Interferon

Dosage (duration [months])

Route

No. of patients

Clinical type

Results

Camenga et al. (1986) Johnson et al. (1990)

db. pc

0'2

SC

98

p-ser-17

SC

30

RR. RP, CP RR

No difference

db , pc

Multicentre IFNp-ser-17 (in progress) Mult icentre 'Bioferon' (in progress)

db, pc

p-ser-17

SC

350

RR

Lower attack rate in IFN group

db. pc

p

2.MU 3 times per week [121 Various doses , then 45MU 3 times per week [361 45MU or 9MU every other day [241 6 MU/week [241

1M

288

RR

In progress

Panitch et al. (1987b,c)

sb

IV

18

RR

Increased attacks during treatment

15 OOOU to 15MU twice per week

Lower attack rate in IFN group

[11 Abbreviations: IFN == interferon; MU == million units ; IV == intravenous; SC == subcutaneous; 1M == intramuscular; RR == relapsingremitting; RP == relaps ing progressive; CP == chronic progressive; sb == single-blind; db == double-blind; pc == placebo-controlled.

ciated. However, they provided the background for further clinical trials based on more recent developments in our understanding of the immunological properties of the IFNs. The following discussion emphasises those immunoregulatory functions thought to be important in MS, the dominant themes being the probable role of IFNI' as an activator of disease activity and the suppressive or inhibitory effects of IFN~ on immune activation by IFNI'. The IFNs can modulate the immune system through induction of MHC class I (HLA-A, Band C) and class II (HLA-DR, DQ and DP) cell surface molecules, which are essential for basic immune functions such as self-nonself discrimination and antigen presentation to T lymphocytes. Antigens associated with class I molecules are recognised by CD8+ (suppressor/cytotoxic) T cells, whereas antigens associated with class II molecules are recognised by CD4+ (helper/inducer) T cells. Proliferation of T lymphocytes is proportional to the density of class II molecules on antigen-presenting cells (Matis et al. 1985). Class I molecules are constitutively expressed on nearly all cells, and all 3 types of IFN augment their expression. Class II

molecules are normally restricted to monocytes, macrophages, B lymphocytes, activated T lymphocytes and dendritic cells (McFarlin 1988). The principal regulator of their expression on these cells, and their induction on other cell types is IFNI' (Basham & Merigan 1983; Sztein et al. 1984). There is little or no expression of class II antigens in normal brain; however, exposure to IFNI' induces them on astrocytes, microglia and endothelial cells, which can then effectively present antigens (Fierz et al. 1985; McCarron et al. 1986), or act as targets for cytotoxic T cells (Dhib-Jalbut et al. 1990). Although the mechanism by which interaction of IFNI' with its receptor initiates gene transcription is not fully understood, intracellular proteins induced by IFNI', such as CRG-2 in mice and IP-I0 in humans, may mediate induction of MHC molecules (Vanguri & Farber 1990). IFNI' also activates macrophages, which act as effector cells in demyelination (Bever & Whitaker 1985; Prineas & Graham 1981 ) and synthesise proteinases that can degrade myelin proteins (Bever 1991). Induction of adhesion molecules that regulate homing of lymphocytes to sites of inflammation, and may facilitate their entry into the CNS (Male et al. 1990),

Interferons in Multiple Sclerosis

is another function of IFN'Y that could affect MS activity. The immune-activating effects of IFN'Y are modulated by other cytokines, For example, tumour necrosis factor (TNF-a), which is found in the CSF of patients with active MS (Sharief & Hentges 1991), and which may itself participate in demyelination (Robbins et al. 1987; Selmaj & Raine 1988), can augment the effect of IFN'Y on induction of class II molecules. Interleukin-I, corticosteroids, prostaglandins, e-fetoprotein, transforming growth factor {J (TGF{J), and noradrenaline (norepinephrine) have all been shown to downregulate class II antigen expression in various experimental systems (Cowan et al. 1991; Frohman et al. 1988; Racke et al. 1991; Ransohoff 1989). Interestingly, If-Nv-stimulated class II molecules can also be downregulated by IFNa and IFN{J. This was first shown in murine macrophages (lnaba et al. 1986; Ling et al. 1985), and later in cultured human cells. When cultured human glioblastoma cells are incubated with IFNs 'Y and {J simultaneously, IFN{J (100 Uzml) inhibits expression of HLA-DR (Joseph et al. 1988). A similar effect of IFN{J has been demonstrated in cultured adult human astrocytes (Barna et al. 1989), which express HLA-DR only after exposure to JFN'Y. The addition of IFN{J results in significant, though not total , inhibition of HLA-DR expression. IFN{J acts by interfering with transcription of class II-specific mRNA, both in the murine macrophage system (Fertsch et al. 1987) and in human astrocytoma cell lines (Ransohoff et al. 1991). The effect ofIFN{J on human monocytes or macrophages is less clearly defined , but may be more relevant to MS, since there is evidence to suggest that perivascular microglia , probably derived from circulating monocytes, are the primary antigen-presenting cells in the CNS (Hayes et al. 1987; Hickey & Kimura 1988; Sedgwick et al. 1991). We found that when isolated monocytes are incubated with IFN{J and IFN'Y simultaneously, there is a partial inhibitory effect on IFN'Y-induced HLADR expression, and a more pronounced effect on HLA-DQ expression (Pan itch et al. 1989). Others, using a different IFN{J preparation, have been unable to confirm these results (Rudick et al. 1989).

953

A suppressive effect of recombinant IFN{J on synthesis of IFN 'Y by cultured peripheral blood mononuclear cells (presumably T lymphocytes) has now been described in 2 laboratories (Noronha et al. 1991; Panitch et al. 1987a), and may be an important factor in the immunoregulation of MS. Although it is generally accepted that IFNs do not penetrate the blood-brain barrier effectively, systemic administration of IFN'Y to MS patients can precipitate acute exacerbations (Panitch et al. 1987a). IFN'Y is also present, in association with class II antigens, in active MS plaques (Traugott & Lebon 1988a). Downregulation of IFN'Y synthesis by systemic IFN{J could thus be an effective means of preventing or modulating the severity of MS attacks. The most widely accepted systemic immunoregulatory defect in MS is the presence of abnormal suppressor T cell function in patients undergoing acute exacerbations or chronic progression (Antel et al. 1978, 1986). It has been recognised for some time that the effects of IFNs a and {J on suppressor T cell activity are distinct from those ofIFN'Y (Aune & Pierce 1982; Frasca et al. 1988; Schnaper et al. 1983), but only recently was this documented in the setting of human autoimmune disease. Noronha et al. (1990) reported that recombinant IFN{J improves suppressor function of T cells from both MS patients and control subjects, and we have confirmed their results (Panitch et al. 1992b). The biological significance of nonspecific suppressor cell activity is problematic because suppression is measured by reduction in proliferation of responder cells exposed to concanavalin-A or antiCD3 (OKT3) monoclonal antibody, or by changes in IgG secretion in cell culture supernatants, neither of which is likely to be a critical factor in the pathogenesis MS. We recently found, however, that IFN{J can modify the regulatory function of T cells on IFN'Y synthesis, and that IFN'Y synthesis is a sensitive indicator of suppressor T cell activity (Panitch et al. 1991). Regulation ofIFN'Y synthesis by suppressor T cells is not a novel observation (Seki et al. 1986). In fact, Swanborg and colleagues (Karpus & Swanborg 1989; McDonald & Swanborg 1988) showed that suppressor cells isolated

954

from rats after recovery from EAE could inhibit production of IFN')' by EAE-effectorcells. The possible implications for MS or other autoimmune diseases have not yet been investigated further. The ability to induce suppressor function in vivo would provide additional justification for the use of IFN,8 as a therapeutic agent in MS.

5. Additional Evidence for a Pathogenic Role of Interferon Gamma in MS Regulation of antigen presentation by class II MHC molecules (Ia in animals) is an essential feature of EAE, the best available animal model of MS. la antigens can be identified on inflammatory, endothelial and microglia cells in the CNS, although la expression on astrocytes is rare (Hickey & Kimura 1988; Matsumoto et al. 1986; Sobel et al. 1984). Treatment of mice with specific anti-Ia monoclonal antibodies inhibits the disease (Sriram & Steinman 1983; Steinman et al. 1981). In vitro, la induction on astrocytes (Fierz et al. 1985) or endothelial cells (McCarron et al. 1986) by IFN')' permits them to present MBP to sensitised T cells. However, continuous intravenous infusion of IFN')' in normal Lewis rats induces la expression on microglia and ependymal cells (Steiniger & van der Meide 1988), suggesting that these may be more important as antigen-presenting cells in vivo. The significance of microglia in antigen presentation within the CNS has also been recognised in other experimental systems (Frei et al. 1987; Hayes et al. 1987; Hickey & Kimura 1988). IFN')' has clearly been shown to potentiate 2 immune-mediated experimental diseases, collagen-induced arthritis (Cooper et al. 1988) and autoimmune neuritis (Hartung et al. 1990). However, it can have apparently conflicting effects on EAE depending on the experimental system used. Adoptive transfer of sensitised lymphocytes after incubation with IFN')' enhances EAE (Racke et al. 1990), but intraventricular administration of IFN')' to rats can inhibit the disease (Voorthuis et al. 1990), and systemic administration of antibody to IFN')' appears to exacerbate it (Billiau et al. 1988).

Drugs 44 (6) 1992

The effects of IFN,8 are somewhat less confusing. EAE in Lewis rats can be prevented by systemic IFN,8 (Abreu 1982), and adoptive transfer of EAE is inhibited by preincubating MBP-sensitised lymphocytes with IFN,8 (Abreu 1985). It is likely, though unproven, that downregulation of la expression is involved . These findings suggest that IFN,8and IFN')' can both modulate disease activity in EAE, but a variety of other regulatory processes are also likely to be involved in determ ining the final outcome. Immunohistochemical studies also suggest that IFNs playa role in the pathogenesis of MS. Patterns of class I and class II antigen expression reportedly correlate with lesion development, class I molecules appearing on astrocytes near inflammatory infiltrates of T cells in acute plaques, and class II (HLA-DR) molecules on astrocytes and macrophages in more chronic active plaques (Traugott 1987). In another study, all 3 types of IFN were identified, with the distribution of IFN')' coinciding with class II antigens at sites of active demyelination (Traugott & Lebon 1988a). Astrocytes expressing IFN')' were especially abundant, even in the absence of inflammatory cells, in white matter of MS patients who had died with pneumonia (Traugott & Lebon 1988b), suggesting that intercurrent infection could activate cells in the CNS to synthesise IFN')', leading to increased expression of class II antigens, presentation of myelin antigens to sensitised T cells, and activation of the disease process. TNF-a, which acts synergistically with IFN')' to induce class II antigens, has also been localised immunohistochemically in MS brain tissue (Hofman et al. 1990). A triggering effect of infections was documented more explicitly by Sibley et al. (1985), who followed 170 patients with MS and 134 healthy controls for an average of 5.3 years to compare the frequency of common, presumably viral, respiratory and enteric infections in the 2 groups. During periods 'at risk' (from 2 weeks before to 5 weeks after onset of infection), the exacerbation rate was nearly 3 times the rate in periods when the risk factor was absent. It is tempting to speculate that IFN')' or other cytokines induced by viral infection

Interferons in Multiple Sclerosis

may have mediated the triggering effect. Our own recent data tend to corroborate this study (Panitch et al. 1992a). In 30 patients with relapsing-remitting MS followed prospectively for 60 weeks, 68% of all acute attacks occurred within 6 weeks of upper respiratory infections, and 32% of such infections were accompanied or followed by attacks. These figures are considerably higher than those previously reported, but the same trend is apparent, indicating a triggering effect on MS activity. Studies of IFN synthesis in MS initially showed reduced levels of IFNa (Neighbour et al. 1981) and IFN-y (Vervliet et al. 1984, Vervliet et al. 1983), associated with abnormally low natural killer cell activity. These findings prompted early attempts to treat MS by replacing the missing IFN . It soon became clear, however, that many MS patients had normal natural killer cell function (Rice et al. 1983; Santoli et al. 1981), and activation of natural killer cells by IFN had no effect on the clinical course of the disease (Hirsch & Johnson 1985; Rice et al. 1983). Hirsch et al. (1985), using a sensitive radioimmunoassay, found that peripheral blood mononuclear cells from relapsing-remitting MS patients synthesised greater amounts of IFN-y than control subjects, and Beck et al. (1988) reported increased levels of IFN-y and TNF-a produced by mitogen-stimulated peripheral blood cells preceding acute exacerbations, speculating that these cytokines represented a triggering mechanism for MS attacks. Other investigators have searched for IFNs and other cytokines in MS serum or CSF, with negative or equivocal results (Franciotta et al. 1989; Hauser et al. 1990; Salonen 1983). In one study, however, very high levels ofTNF-a, that seemed to correlate remarkably well with disease severity, were reported in the CSF of chronic progressive MS patients (Sharief & Hentges 1991). TNF-a, which acts synergistically with IFN-y to upregulate class II antigen expression on astrocytes (Benveniste et al. 1989; Vidovic et al. 1990), may be an important cofactor in the immunoregulation of MS. The pivotal event establishing the importance of If'Nv as a mediator ofMS activity was an actual clinical trial of recombinant IFN-y (Panitch et al.

955

1987a, b). It was recognised in the early 1980s that IFNy had the ability to stimulate HLA-DR expression (Basham & Merigan 1983; Kelley et al. 1984) and to activate macrophages (Black et al. 1987; Murray et al. 1987;' Nathan et al. 1983), both of which could be potentially hazardous in MS. On the other hand , investigators had reported that synthesis ofIFN-y was deficient in cultures of mitogenstimulated peripheral blood leucocytes from MS patients (Vervliet et al. 1983, 1984). The only way to resolve the issue definitively was to undertake a carefully monitored pilot study to assess the safety of IFN-y and its effect on disease activity. 18 patients with clinically definite or laboratory-supported relapsing-remitting MS were randomised to receive either 1, 30 or 1000/Lg (corresponding to 15 000, 450 000, or 15 million U, respectively) of IFN-y (Immuneron'P, Biogen Research Corporation) by intravenous infusion twice a week for 4 weeks. During treatment, 7 of the 18 patients had exacerbations, all of which were confined to signs and symptoms experienced during previous attacks. Although MRI with gadolinium enhancement was not available during this trial, the clinical findings suggestedthat IFN-y or IFN-y-activated cells had infiltrated the CNS in areas of previous bloodbrain barrier disruption, and reactivated individual lesions. Concurrent immunological studies showed dose-dependent induction of HLA-DR on peripheral blood monocytes during the course of IFN treatment, suggesting a causal relationship to the triggering of attacks (Panitch et al. 1987b). The study was discontinued because of the unacceptably high relapse rate, but it provided strong support for clinical trials of other agents, including IFNs a and (j, which can inhibit the effects of'If'Nv on the immune system. Recently Bever et al. (1991) published their immunological findings in 18 chronic progressive MS patients treated with the IFN inducer poly-ICLC. Peak levels of IFN were detected in serum 12 hours after intravenous infusion of the drug, which induced IFNs a and {j, but also measurable amounts ofIFN-y. The mean peak level ofIFN-y was 15900 UIL , compared to a mean total IFN level of 177000 U/L. Nevertheless , IFN-y levels did not correlate

956

with clinical worsening; in fact the highest levels were detected in patients who improved or remained stable. It was suggested that the high levels of IFNs a and {3 may have blocked immune activation by IFN-y. Other possibilities were that cortisol induced by poly-ICLC infusion inhibited clinical disease activity (Bever et al. 1988), that the patients were not susceptible to the activating effects of IFN-y, or that disease activity induced in patients with chronic progressive MS and fixed neurological deficits was simply not clinically apparent.

6. Current Therapeutic Trials of Recombinant Interferon Beta IFN{3 was cloned and expressed in bacteria as early as 1980, but proved to be unsuitable for use in clinical trials until it was found that insertion of a serine residue in place of cysteine at position 17 greatly improved stability (Khosrovi 1984). This genetically engineered preparation , known as IFN{3ser-17 or Betaseron'P (now produced by Berlex Laboratories, Inc.), retained the specific activity of natural IFN{3. Testing in cancer patients showed IFN{3-ser-17 to be well tolerated and remarkably free of side effects, even at high doses. However, it was also reported to enhance class II antigen expression on human monocytes (Spear et al. 1987), raising the possibility that, like IFN-y, it might trigger exacerbations of MS. Therefore, a pilot study was undertaken to assess the safety and maximum tolerated dose of IFN{3-ser-17 in MS patients (Johnson et al. 1990). Thirty patients with clinically proven relapsingremitting MS were randomised to receive either 90, 45, 22.5, or 4.5 million U of IFN{3-ser-17, or an equivalent volume of placebo vehicle, by subcutaneous injection 3 times a week. It quickly became apparent that patients did not tolerate the 90 million U dose, but tolerated 45 million U with minimal side effects. After 6 months, all patients except those on placebo were switched to the 45 million U dose, and continued to receive it for a total of 3 years with no adverse reactions and no

Drugs 44 (6) 1992

increase in relapse rate. In the first 6 months, there was a dose-dependent reduction of attacks in the IFN-treated patients . Subsequently, 10 IFN-treated patients (42%), but only 1 placebo-treated patient (17%) remained free of attacks over the 3-year study period. Approximately two-thirds of the patients receiving IFN developed neutralising antibodies , but there was no relationship between antibody titres and clinical efficacy. Many of the participants in this study have now received IFN{3-ser-17 for over 6 years with no ill effects. In a subgroup of patients, IFN-y synthesis by concanavalinA-stimulated peripheral blood lymphocytes was reduced during the first 6 months in proportion to the dose of IFN received, and the effect was then maintained on the 45 million U dose. No significant changes were found in HLA-DR expression on peripheral blood mononuclear cells, despite the fact that IFN{3-ser-17 reduced expression of class II antigens on If-Nv-stimulated monocytes in vitro. Because of these encouraging results, a multicentre, placebo-controlled, double-blind trial was organised in 1988. Over 350 subjects with clinically definite relapsing-remitting MS were enrolled at 12 centres in the US and Canada. All were ambulatory with Kurtzke expanded disability status scale (EDSS) scores (Kurtzke 1983) of 5.5 or less, and a history of 2 or more relapses in the previous 2 years. They were randomised to receive either high dose (45 million U) or low dose (9 million U) IFN{3-ser-17 or placebo by subcutaneous injection every other day for 2 years. Subjects were rated on the EDSS and Functional Systems scales, and on the Scripps Neurologic Rating Scale (Sipe et al. 1984) at each scheduled visit, and at the time of each attack. Serial quantitative MRI scans were performed before treatment, and after the first and second years. At several centres, paraclinical investigations were conducted, including studies of IFN-y synthesis, class II antigen expression, suppressor cell induction, proteinase activity and MBP breakdown products in CSF and urine. One of these studies, an analysis of serum cortisol levels after IFN injection , has already reported no change in patients receiving IFN{3 (Reder et al. 1989). In view of previous studies showing increased serum cor-

Interferons in Multiple Sclerosis

tisol levels after administration of IFNa (Roosth et al. 1986) or poly-ICLC (Bever et al. 1988), this is an important observation because it shows that the clinical effect of systemic IFNfj is not simply a nonspecific one similar to that of corticotrophin. Clinical data for the first 2 years of the study have been analysed and a significant result favouring IFNfj treatment was found . The data are not yet available to the investigators, who remain blinded , but an application was submitted to the Food and Drug Administration in June 1992 requesting approval of IFNfj-ser-17 for the treatment of relapsing-remitting MS. Meanwhile, the patients enrolled in the study will remain on treatment to permit accumulation of additional information on safety and efficacy. This study has the potential to provide information not only on reduction of exacerbation rates, but also on the effect of treatment on progressive neurological impairment over an extended period of time. While the IFNfj-ser-17 study was in progress, Jacobs and colleagues designed and initiated a multicentre clinical trial of a different variety of recombinant IFNfj, Bioferon'P, a glycosylated molecule produced in a mammalian cell line rather than in bacteria (Jacobs, personal communication). Although previous studies by these investigators demonstrated a degree of efficacy for natural IFNfj administered intrathecally, further pursuit of the intrathecal route was deemed impractical. In the current trial, 6 million U ofIFN, or a placebo, will be administered weekly by intramuscular injection for 2 years to patients with relapsing-remitting MS. They will then be observed periodically for an additional 2 years, with the principal outcome measures being changes in EDSS scores and differences in exacerbation rates between the IFN and placebo-treated groups. Many of the features of this study complement those of the IFNfj-ser-17 trial described above, and the results of both are awaited with interest. These 2 well-designed studies should add substantially to our knowledge of the role of the IFN system in the pathogenesis of MS, and are likely to be considered definitive with respect to the therapeutic value of IFNfj in relapsing-remitting MS.

957

7. Conclusions MS characteristically strikes young people in their most productive years, gradually resulting in severeneurological deficits. There is a pressing need for therapeutic agents with minimal toxicity that can be administered early in the disease, that can be given indefinitely, and that prevent both exacerbations and cumulative disability. It has been said that prevention of MS exacerbations is not a worthwhile goal of treatment, since the underlying disease continues to progress. Because of this, some investigators doubt the value of IFN therapy. However, a recent study of the natural history of MS (Weinshenker et al. 1989) showed that exacerbation rates early in the clinical course were directly proportional to ultimate disability. Evidence also exists that relapsing-remitting MS may differ not only clinically, but also genetically (Olerup et al. 1989)and pathologically(Thompson et al. 1991) from MS which is progressive from the outset. Thus, the responses to IFN therapy in these settings may also be different. If IFNfj is well tolerated by patients and proves to have a beneficial biological effect on MS by reducing the attack rate, the burden of proof would seem to rest with those who claim that IFNfj does not affect the ultimate outcome. Meanwhile, it may well become a new standard against which other therapeutic agents are measured . In this review we have not considered recent developments in specific immunotherapy of MS, such as peptide vaccines based on preferential usage of T cell receptors (Kotzin et al. 1991). These exciting concepts are in their infancy, and will require years of further research and development before large scale clinical trials can be considered. In the realm of nonspecific immunotherapy, recombinant IFNfj currently holds the greatest promise of any available agent for early relapsingremitting MS, and may become the first new agent formally approved for treatment of MS since corticotrophin and corticosteroids in the early 1970s. However, it is unlikely that any type or dose of IFN will be the final answer to the problem. Combination therapies with corticosteroids , other im-

Drugs 44 (6) 1992

958

munosuppressive agents, inhibitors ofTNF-a (Selmaj et aJ. 1991), or immunosuppressive cytokines such as TGF-13 (Johns et aJ. 1991 ; Racke et aJ. 1991), must also be considered. Such forms of treatment, based on the cancer chemotherapy model, may hold more promise than the use of IFNI3 alone.

References Abreu SL. Suppression of experimental allergic encephalomyelit is by interferon. Immunological Communications II: 1-7, 1982 Abreu SL. Interferon in experimental autoimmune encephalomyelitis (EAE): effects of exogenous interferon on the antigen-enhanced adoptive transfer of EAE. International Archives of Allergy and Applied Immunology 76: 302307,1985 Allegretta M, Nicklas JA, Sriram S, Albert ini RJ. T cells respons ive to myelin basic protein in patients with multiple sclerosis . Science 247: 718-721, 1990 Antel JP , Amason BGW, Medof ME. Suppressor cell function in multiple sclerosis: correlation with disease activity. Annals of Neurology 6: 469-473, 1978 Antel JP , Bania MB, Reder A, Cashman N. Activated suppressor cell dysfunction in progressive mul tiple sclerosis. Journal of Immunology 137: 137-141, 1986 Aune TM , Pierce CWoActivation ofa suppressor T-cell pathway by interferon. Proceed ings of the Nat ional Academ y of Sciences USA 79: 3808-3812, 1982 AUSTIMS Research Group. Interferon-a and transfer factor in the treatment of multiple sclerosis: a double-blind , placebo-eontrolled trial. Journal of Neurology, Neurosurgery, and Psychiatry 52: 566-574, 1989 Barna BP, Chou SM, Jacobs B, Yen-Lieberman B, Ransohoff RM. Interferon-s impairs induction of HLA-DR ant igen expression in cultured adult human astrocytes. Journal of Neuroimmunology 23: 45-53, 1989 Basham TY , Merigan TC. Recombinant interferon-gamma increases HLA-DR synthesis and expression. Journal of Immunology 130: 1492-1494, 1983 Beck J, Rondot P, Catinot L, Falcoff E, Kirchner H, et al. Increased production of interferon gamma and tumor necrosis factor precedes clinical manifestation in multiple sclerosis: do cytokines trigger off exacerbations? Acta Neurologica Scandinavica 78: 318-323, 1988 Ben-Nun A, Liblau RS, Cohen L, Lehmann D, TournierLasserve E, et al. Restricted T-cell receptor tl gene usage by myelin basic protein-specific T-cell clones in multiple sclerosis: predominant genes vary in individuals. Proceedings of the National Academy of Sciences, USA 88: 24662470, 1991 Benveniste EN, Sparacio SM, Bethea JR . Tumor necrosis . factor-a enhances interferon-s mediated class II ant igen expression on astrocytes. Journal of Neuro immunology 25: 209-219, 1989 Bever CT. Cathepsin B in peripheral blood mononuclear cells of MS patients. Abstract. Transactions of the American Society for Neurochemistry 22: 208, 1991 Bever CT, McFarland HF, Levy HB, McFarlin DE. Cortisol

induction by poly ICLC: implications for clinical trials of interfero n. Annals of Neurolog y 23: 196-199, 1988 Bever CT, Panitch HS, Levy HB, McFarlin DE, Johnson KP . Gamma-interferon induction in patients with chronic progressive MS. Neurology 41: 1124-1127, 1991 Bever CT, Salazar AM, Neely E, Ferraraccio BE, Rose JW , et al. Preliminary trial of poly ICLC in chronic progressive multiple sclerosis. Neurology 36: 494-498, 1986 Bever CT, Wh itaker IN. Proteinases in inflammatory demyelinating disease . Springer Seminars in Immunopathology 8: 235-250, 1985 Billiau A. Redefin ing interferon: the interferon-like antiviral effects of certain cytokines (interleukin-I , interferon-s, interferon-v) may be indirect or side effects. Antiviral Research 8: 55-70, 1987 Billiau A, Heremans H, Vandekerckhove F, Dijkmans R, Sobis H, et al. Enhancement of experimental allergic encephalomyelitis in mice by antibodies against IFN'Y. Journal of Immunology 140: 1506-1510, 1988 Black CM, Catterall JR , Remington JS. In vivo and in vitro activation of alveolar macrophages by recombinant interferonv, Journal of Immunology 138: 491-495,1987 Booss J, Kim JH . Evidence for a viral etiology of multiple sclerosis. In Cook SD (Ed.) Handbook of multiple sclerosis, pp. 41-61, Marcel Dekker, New York, 1990 Camenga DL, Johnson KP, Alter M, Engelhardt CD, Fishman PS, et al. Systemic recombinant a-2 interferon therap y in relapsing multiple sclerosis. Archi ves of Neurolog y 43: 1239-1246, 1986 Cantell K, Hirvonen S. Large-scale production of human leucocyte interferon containing 108 units per ml. Journal of General Virology 39: 541-543, 1978 Cook SD, Dowling Pc. Multiple sclerosis and viruses: an overview. Neurolog y 30 (2): 80-91, 1980 Cooper SM, Sriram S, Ranges GE. Suppression of murine collagen-induced arthritis with monoclonal ant i-Ia antibodies and augmentation with IFN 'Y . Journal of Immunology 141: 1958-1962, 1988 Cowan E, Pierce M, Dhib-Jalbut S. Interleukin-I downregulates HLA class II expression on a human glioblastoma cell line. Journal of Neuroimmunology 33: 17-28, 1991 De Maeyer E, De Maeyer-Guignard J. Chapter 13: Interactions of interferons, tumor necrosis factor, interleukin-I, and interleukin-2 as part of the cytokine network . In De Maeyer & DeMaeyer-Guignard (Eds) Interferons and other regulatory cytokines , pp. 288-333, John Wiley & Sons, New York, 1988a De Maeyer E, De Maeyer-Guignard J. Chapter 2: The interferon gene family . In De Maeyer & De Maeyer-Guignard (Eds) Interferons and other regulatory cytokines, pp. 5-38, John Wiley & Sons, New York, 1988b De Maeyer E, De Maeyer-Guignard J. Chapter 3: Induction of IFN-a and IFN-tl. In De Maeyer & De Maeyer-Guignard (Eds) Interferons and other regulatory cytokines, pp. 39-66, John Wiley & Sons, New York, 1988c Dhib-Jalbut S, Kufta CV, flerlage M, Shimojo N, McFarland HF. Adult human glial cells can present target antigen to HLA-rest ricted cytotoxic T-cells. Journal of Neuroimmunology 29: 203-211, 1990 Dhib-Jalbut S, McFarl in DE. Immunology of multiple sclerosis. Annals of Allergy 64: 433-444, 1989 Fertsch D, Schoenberg DR, Germain RN , Tou JYL , Vogel

Interferons in Multiple Sclerosis

SN. Induction of macrophage Ia antigen expression by rlFN'Y and down-regulation by IFN-a/I'l and dexamethasone are mediated by changes in steady-state levels of Ia mRNA. Journal of Immunology 139: 244-249, 1987 Fierz W, Endler B, Reske K, Wekerle H, Fontana A. Astrocytes as antigen-presenting cells. I. Induction of Ia antigen expression in astrocytes by T cells via immune interferon and its effect on antigen presentation. Journal of Immunology 134: 3785-3793, 1985 Fleischmann WR, Ramamurthy V, Stanton GJ, Baron S, Dianzani F. Interferon : mode of action and clinical applications. In Smith R (Ed.) Interferon treatment..of neurologic disorders, pp. 1-42, Marcel Dekker, New York, 1988 Franciotta DM, Grimaldi LME,Martino GV, Piccolo G, Bergamaschi R, et al. Tumor necrosis factor in serum and cerebrospinal fluid of patients with multiple sclerosis. Annals of Neurology 26: 787-789, 1989 Frasca D, Adorini L, Landolfo S, Doria G. Enhancement of suppressor T cell activity by injection of anti-IFNI' monoclonal antibody . Journal of Immunology 140: 4103-4107, 1988 Freedman M, Ruijs TCG, Selin LK, Antel JP. Peripheral blood I' T cells lyse fresh human brain-derived oligodendrocytes. Annals of Neurology 30: 794-800, 1991 Frei K, Seipl C, Groscurth P, Bodmer S, Schwerdel C, et al. Antigen presentation and tumor cytotoxicity by interferonv-treated microglial cells. European Journal of Immunology 17: 1271-1278, 1987 Frohman EM, Vayavegula B, Gupta S, van den Noort S. Norepinephrine inhibits I' interferon-induced major histocompatibil ity class II (la) antigen expression on cultured astrocytes via 1'l2-adrenergic signal transduction mechanisms. Proceedings of the National Academy of Sciences USA 85: 1292-1296, 1988 Gonzalez-Scarano F, Grossman RI, Galetta S, Atlas SW, Silberberg DH. Multiple sclerosis disease activity correlates with gadolinium -enhanced magnetic resonance imaging. Annals of Neurology 21: 300-106, 1987 HabifDV, Lipton R, Cantell K. Interferon crosses the bloodcerebrospinal fluid barrier in monkeys. Proceedings of the Society for Experimental Biology and Medicine 149: 287289,1975 Hafler DA, Fox DA, Manning ME, Schlossman SF, Reinherz EL, et al. In vivo activated T lymphocytes in the peripheral blood and cerebrospinal fluid of patients with multiple sclerosis. New England Journal of Medicine 312: 14051411, 1985 Hafler DA, Weiner HL. MS: a CNS and systemic autoimmune disease. Immunology Today 10: 104-107, 1989 Harris JO, Frank JA, Patronas N, McFarlin DE, McFarland HF. Serial gadolinium-enhanced magnetic resonance imaging scans in patients with early, relapsing-remitting multiple sclerosis: implications for clinical trials and natural history. Annals of Neurology 29: 548-555, 1991 Hartung H-P, Schafer B, van der Meide FH, Fierz W, Heininger K, et al. The role of interferon-gamma in the pathogenesis of experimental autoimmune disease of the peripheral nervous system. Annals of Neurology 27: 247-257, 1990 Hauser SL, Doolittle TH, Lincoln R, Brown RH, Dinarello CA. Cytokine accumulations in CSF of multiple sclerosis patients : frequent detection of interleukin -I and tumor ne-

959

crosis factor but not interleukin-6. Neurology 40: 17351739, 1990 Hayes GM, Woodroofe MN, Cuzner ML. Microglia are the major cell type expressing MHC class II in human white matter. Journal of the NeurologicalSciences80: 25-37,1987 Hickey WF, Kimura H. Perivascular microglial cells of the CNS are bone marrow derived and present antigen in vivo. Science 239: 290-292, 1988 Hirsch RL, Johnson KP. Natural killer cell activity in multiple sclerosis patients treated with recombinant interferon-oj, Clinical Immunology and Immunopathology 37: 236-244, 1985 Hirsch RL, Johnson KP, Camenga DL. The placebo effect during a double blind trial of recombinant alpha-2 interferon in multiple sclerosis patients : immunological and clinical findings. International Journal of Neuroscience 39: 189-196, 1988 Hirsch RL, Panitch HS, Johnson KP. Lymphocytes from multiple sclerosis patients produce elevated levels of gamma interferon in vitro . Journal of Clinical Immunology 5: 386-389, 1985 Hofman FM, Hinton DR, Johnson K, E MJ. Tumor necrosis factor identified in multiple sclerosis brain. Journal of Experimental Medicine 170: 607-612, 1990 Inaba K, Kitaura M, Kato T, Watanabe Y, Kawade Y, et al. Contrasting effect of Ol/I'l and interferons on expression of macrophage Ia antigens. Journal of Experimental Medicine 163: 1030-1035, 1986 Isaacs A, Lindenmann J. Virus interference. I. The interferon. Proceedings of the Royal Society of London (Biology) 147: 258-267, 1957 Jacobs L, O'Malley J, Freeman A, Ekes R. Intrathecal interferon reduces exacerbations of multiple sclerosis. Science 214: 1026-1028, 1981 Jacobs L, Salazar AM, Herndon R, Reese PA, Freeman A, et al. Multicentre double-blind study of effect of intrathecally adm inistered natural human fibroblast interferon on exacerbations of multiple sclerosis. Lancet 2: 1411-1413, 1986 Jacobs L, Salazar AM, Herndon R, Reese PA, Freeman A, et al. Intrathecally administered natural human fibroblast interferon reduces exacerbations of multiple sclerosis: Results of a multicenter , double-blinded study. Archives of Neurology 44: 589-595, 1987 Johns LD, Flanders KC, Ranges GE, Sriram S. Successful treatment of experimental allergic encephalomyelitis with transforming growth factor-sf. Journal of Immunology 147: 1792-1796, 1991 Johnson KP, Knobler RL, Greenstein 11, Panitch HS, Lublin FD, et al. Recombinant human interferon beta treatment of relapsing-remitting multiple sclerosis: pilot study results. Abstract. Neurology 40 (Suppl. I): 261, 1990 Johnson KP, Panitch HS. Interferon therapy of multiple sclerosis. In Cook SD (Ed.) Handbook of multiple sclerosis, pp. 455-468, Marcel Dekker, New York, 1990 Joseph J, Knobler RL, D'Imperio C, Lublin FD. Down-regulation of interferon-v -induced class II expression on human glioma cells by recombinant interferon-s effects of dosage treatment schedule. Journal of Neuroimmunolog y 20: 39-44, 1988 Karpus WJ, Swanberg RH. CD4+ suppressor cells differentially affect the production of IFNI' by effector cells of

960

experimental autoimmune encephalom yelitis. Journal of Immunology 143: 3492-3497, 1989 Kastrukoff LF, Oger JJ , Hashimoto SA, Sacks SL, Li DK, et al. Systemic Iymphoblastoid interferon therapy in chron ic progressive multiple sclerosis. I. Clinical and MRI evaluation . Neurology 40: 479-486, 1990 Kastrukoff LF, Oger JJ , Tourtellotte WW, Sacks SL, Berkowitz J, et al. Systemic Iymphoblastoid interferon therapy in chronic progressive multiple sclerosis. II. Immunologic evaluation . Neurology 41: 1936-1941, 1991 Kelley VE, Fiers W, Strom TE. Cloned human interferon')', but not interferon-s or a, induces expression of HLA-DR determinants by fetal monocytes and myeloid leukemic cell lines. Journal of Immunology 132: 240-245, 1984 Kermode AG, Tofts PS, Thompson AJ, MacManus DG, Rudge P, et al. Heterogerieity of blood-brain barrier changes in multiple sclerosis:an MRI study with gadolinium-DTPA enhancement. Neurology 40: 229-235, 1990 Khosrovi B. The production, characterization, and testing of a modified recombinant human interferon beta. In Zoon K et al. (Eds) Interferon : research, clinical application, and regulatory consideration, pp. 89-99, Elsevier Science Publishing, Amsterdam, 1984 Knobler RL, Panitch HS, Braheny SL. Controlled clinical trial of systemic alpha interferon in multiple sclerosis. Neurology 34: 1273-1279, 1984 Kotzin BL, Karuturi S, Chou YK, Lafferty J, Forrester JM, et al. Preferential T-cell receptor chain variable gene use in myelin basic protein-reactive T-cell clones from patients with multiple sclerosis. Proceedings of the National Academy of Sciences USA 88: 9161-9165,1991 Krown SE. AIDS and Kaposi's sarcoma: Interferons in pathogenesis and treatment. In Gresser I (Ed.) Interferon 7, pp. 185-211, Academic Press, London, 1986 Kurtzke JF. Rating neurologic impairment in multiple sclerosis: an expanded disability status scale (EDSS). Neurology 33: 1444-1452, 1983 Ling PD, Warren MK, Vogel SN. Antagonistic effect of interferon-d on the interferon-v-induced expression of Ia antigen in murine macrophages. Journal of Immunology 135: 1857-1863, 1985 Male D, Pryce G, Hughes C, Lantos P. Lymphocyte migration into brain modelled in vitro : control by lymphocyte activation, cytokines, and antigen. Cellular Immunology 127: I-II, 1990 Martin R, Jaraquemada D, Flerlage M, Richert J, Whitaker, et al. Fine specificity and HLA restriction of myelin basic protein-specificT cell lines from multiple sclerosis patients and healthy individuals. Journal ofImmunology 145: 540548, 1990 Matis LA, Glimcher LH, Paul WE, Schwartz RH. Magnitude of response of histocompatibility restricted T-cell clones is a function of the product of the concentrations of antigen and Ia molecules. Proceedings of the National Academy of Sciences USA 80: 60I0-6023, 1985 Matsumoto Y, Hara N, Tanaka R, Fujiwara M. Immunohistochemical analysis of the rat central nervous system during experimental allergic encephalomyelitis , with special reference to Ia-positive cells with dendritic morphology. Journal ofImmunology 136: 3668-3676, 1986 McCarron RM, Spatz M, Kempski 0 , Hogan RN, Muehl L, et al. Interaction between myelin basic protein-sensitized

Drugs 44 (6) 1992

T lymphocytes and murine cerebral vascular endothel ial cells. Journal of Immunolog y 137: 3428-3435, 1986 McDonald AH, Swanborg RH. Antigen-specific inhibition of immune interferon production by suppressor cells of autoimmune encephalom yelitis. Journal of Immunology 140: 1132-1138,1988 McFarlin DE. Immunogenet ics in relation to neurologic disease. Allergy Clinics of North America 8: 201-212, 1988 McFarlin DE, McFarland HF. Multiple sclerosis. New England Journal of Medicine 307: 1183-1188, 1246-1251, 1982 Merrill J, Jondal M, Seeley J, Ullberg M, Siden A. Decreased NK killing in patients with multiple sclerosis: an analysis at the level of the single effector cell in peripheral blood and cerebrospinal fluid in relation to the activity of the disease. Clinical and Experimental Immunology 47: 419430, 1982 Meyers CA, Scheibel RS, Forman AD. Persistent neurotoxicity of systemically administered interferon-alpha . Neurology 41: 672-676, 1991 Milanese C, Salmaggi A, La Mantia L, Campi A, Eoli M, et al. Double blind study of intrathecal beta-interferon in multiple sclerosis: clinical and laboratory results. Journal of Neurology, Neurosurgery, and Psychiatry 53: 554-557, 1990 Murray HW, Scavuzzo D, Jacobs JL, Kaplan MH, Libby DM, et al. In vitro and in vivo activation of human mononuclear phagocytes by interferon-s. Journal of Immunology 138: 2457-2462, 1987 Nathan CF, Murray HW, Wiebe ME, Rubin BY. Identification of interferon- s as the Iymphokine that activates human macrophage oxidative metabolism and antimicrobial activity. Journal of Experimental Medicine 158: 670-689, 1983 Neighbour PA, Grayzel AI, Miller AE. Endogenous and interferon-augmented natural killer cell activity of human peripheral blood mononuclear cells in vitro: studies of patients with multiple sclerosis, systemic lupus erythematosus, or rheumatoid arthritis. Clinical and Experimental Immunology 49: 11-21 , 1982 Neighbour PA, Miller AE, Bloom BR. Interferon responses ofleucocytes in multiple sclerosis. Neurology 31: 561-566, 1981 Noronha A, Toscas A, Jensen MA. Interferon beta augments suppressor cell function in multiple sclerosis. Annals of Neurology 27: 207-210, 1990 Noronha A, Toscas A, Jensen MA. IFN-beta down-regulates IFN-gamma production by activated T cells in MS. Abstract. Neurology 41 (Suppl. I): 219,1991 O'Gorrnan MR, Oger JJ, Kastrukoff LF. Reduction of immunoglobulin G secretion in vitro following long term Iymphoblastoid interferon (Wellferon) treatment in multiple sclerosis patients. Clinical and Experimental Immunology 67: 66-75, 1987 Oksenberg JR , Stuart S, Begovich AB, Bell RB, Erlich HA, et al. Limited heterogeneity of rearranged T-cell receptor Va transcripts in brains of multiple sclerosis patients. Nature 345: 344-346, 1990 Olerup 0 , Hillert J, Fredrikson S, Olsson T, Kam-Hansen S, et al. Primaril y chronic progressive and relapsing/remitting multiple sclerosis: Two immunogenetically distinct disease entit ies. Proceedings of the National Academy of Sciences USA 86: 7113-7117, 1989

Interferons in Multiple Sclerosis

Panitch HS, Bever CT, Katz E, Johnson KP. Upper respiratory tract infections trigger attacks of multiple sclerosis in patients treated with interferon-d. Abstract. Journal of Neuroimmunology, in press, 1992a Panitch HS, Folus JS, Johnson KP. Recombinant beta interferon inhib its gamma interferon production in multiple sclerosis. Abstract. Annals of Neurology 22: 139, 1987a Panitch HS, Folus JS, Johnson KP. Beta interferon prevents HLA class II antigen induction by gamma interferon in MS. Abstract. Neurology 39 (Suppl. I): 171, 1989 Panitch HS, Folus JS, Johnson KP. Activated suppressor cells inhibit synthesis of interferon-v in patients with multiple sclerosis and normal subjects. Abstract. Journal of Neuroimmunology, in press, 1992b Panitch HS, Francis GS, Hooper CJ, Merigan TC, Johnson KP. Serial immunological studies in multiple sclerosis patients treated systemically with human alpha interferon . Annals of Neurology 18: 434-438, 1985 Panitch HS, Gomez-Plascencia J, Norris FH, Cantell K, Smith RA. Remission of subacute sclerosing panencephalitis in patients treated with intra ventricular interferon . Neurology 36: 562-566, 1986 Panitch HS, Hirsch RL, Haley AS, Johnson KP. Exacerbations of multiple sclerosis in patients treated with gamma interferon . Lancet I: 893-895, 1987b Panitch HS, Hirsch RL, Schindler J, Johnson KP. Treatment of multiple sclerosis with gamma interferon : exacerbations associated with activation of the immune system. Neurology 37: 1097-1102, 1987c Pestka S. The purification and manufacture of human interferons. Scientific American 249: 36-43, 1983 Prineas JW, Graham JS. Multiple sclerosis: capping of surface immunoglobulin G on macrophages engaged in myelin breakdown . Annals of Neurology 10: 149-158, 1981 Racke M, Dhib-Jalbut S, McFarland H, McFarlin C. Modification of murine experimental allergic encephalomyelitis with cytokines. Abstract. Annals of Neurology 28: 24, 1990 Racke MK, Dhib-Jalbut S, Cannella B, Albert PS, Raine CS, et al. Prevention and treatment of chronic relapsing experimental allergic encephalomyelitis by transforming growth factor-st . Journal ofImmunology 146: 3012-3017, 1991 Ransohoff RM. Regulation of class II major histocompatibility genes: relation to multiple sclerosis. Research in Immunology 140: 202-207, 1989 Ransohoff RM. Pathogenesis of multiple sclerosis: Relationship to therapeutic strategies. In Rudick RA & Goodkin DE (Eds) Treatment of multiple sclerosis: trial design, results, and future perspectives, pp, 123-133, Springer-Verlag, London , 1992 RansohoffRM, Devajyothi C, Estes ML, Babcock G, Rudick RA, et al. Interferon-a specifically inhibits interferon-induced class II major histocompatibil ity complex gene transcription in a human astrocytoma cell line. Journal of Neuroimmunology 33: 103-112, 1991 Reder AT, Lowy MT, Larocca AT, Colby CB . Subcutaneous IFN-P does not affect serum cortisol levels in multiple sclerosis. Abstract. Journal of Interferon Research 9 (Suppl, 2): S220, 1989 Rice GPA, Casali P, Merigan TC, Oldstone MBA. Natural killer cell activity in patients with multiple sclerosis given

961

a interferon . Annals of Neurology 14: 333-338, 1983 Rice GPA, Woelfel EL, Talbot PI, Braheny SL, Sipe JC, et al. Immunological complications in multiple sclerosis patients receiving interferon. Annals of Neurology 18: 439442, 1985 Richert JR , Robinson ED, Johnson AH, Bergman CA, Dragovic U , et al. Heterogeneity of the T-cell receptor p gene rearrangements generated in myelin basic protein-specific T'-cell clones isolated from a patient with multiple sclerosis. Annals of Neurology 29: 209-306, 1991 Robbins OS, Shirazi Y, Drysdale B-E, Lieberman A, Shin HS, et al. Production of cytotoxic factor for oligodendrocytes by stimulated astrocytes. Journal ofImmunology 139: 2593-2597, 1987 Rohatiner AZS, Farkkila M. Neurotoxicity of interferon therapy. In Smith R (Ed) Interferon treatment of neurologic disorders, pp. 135-143, Marcel Dekker, New York, 1988 Rohat iner AZS, Prior PF, Burton AC, Smith AT, Balkwill FR, et al. Central nervous system toxicity of interferon. British Journal of Cancer 47: 419-422, 1983 Roosth J, Pollard RB, Brown SL, Meyer WJ. Cortisol stimulation by recombinant interferon-co. Journal of Neuroimmunology 12: 311-316, 1986 Rudge P. Does a retrovirally encoded superantigen cause multiple sclerosis? Journal of Neurology, Neurosurgery, and Psychiatry 54: 853-855, 1991 Rudick RA, Barna BP, Ransohoff R, Namey M, Jacobs B. Regulation of HLA-DR expression and mono kine secretion by peripheral blood monocytes in multiple sclerosis. Abstract. Annals of Neurology 26: 150, 1989 Salonen R. CSF and serum interferon in multiple sclerosis: Longitudinal study. Neurology 33: 1604-1606, 1983 Santoli D, Hall W, Kastrukoff L, Lisak RP, Perussia B, et al. Cytotoxic activity and interferon production by lymphocytes from patients with multiple sclerosis. Journal of Immunology 126: 1274-1278, 1981 Schnaper H, Aune TM, Pierce CWo Suppressor T cell activation by human leucocyte interferon. Journal of Immunology 131: 2301-2306, 1983 Sedgwick JD, Mobner R, Schwender S, Ter Meulen V. Major histocompatibility complex expressing non-haematopoietic astroglial cells prime only CD8+ T lymphocytes: Astroglial cells as perpetrators but not initiators of CD4+ T cell responses in the central nervous system. Journal of Experimental Medicine 173: 1235, 1991 Seki H, Taga K, Matsuda A, Uwadana N, Hasui M, et al. Phenotypic and functional characteristics of active suppressor cells against IFN-'Y production in PHA-stimulated cord blood lymphocytes.Journal of Immunology 137: 31583161, 1986 Selmaj K, Raine CS, Cross AH. Anti-tumor necrosis factor therap y abrogates autoimmune demyelination . Annals of Neurology 30: 694-700, 1991 Selmaj KW, Raine CS. Tumor necrosis factor mediates myelin and oligodendrocyte damage in vi/roo Annals of Neurology 23: 339-346, 1988 Sharief MK, Hentges R. Association between tumor necrosis factor-a and disease progression in patients with multiple sclerosis. New England Journal of Medicine 325: 467-472, 1991 Sibley WA, Bamford CR, Clark K. Clinical viral infections

962

and mult iple sclerosis. Lancet I: 1313-1315, 1985 Sipe JC, Knobler RL, Braheny SL, Rice GPA, Panitch HS, et al. A neurologic rating scale (NRS) for use in mult iple sclerosis. Neurology 34: 1368-1372, 1984 Smith RA, Abramsky 0 , Steiner I, Panitch H, Cantell K. The experimental treatment of subacute sclerosing panencephalitis with interferon. In Stewart WE et al. (Eds) The biology of the interferon system 1985, pp. 505-510, Elsevier Science Publishers, Amsterdam, 1986 Smith RA, Norris F, Palmer D. Distribution of alpha interferon in serum and cerebrospinal fluid after systemic administration. Clinical Pharmacology and Therapeutics I: 85-88, 1985 Sobel RA, Blanchette BW, Bhan AK, Colvin RB. The immunopathology of experimental allergic encephalomyelitis. II. Endothelial cellla increases prior to inflammatory cell infiltration. Journal of Immunology 132: 2402-2407, 1984 Spear GT, Paulnock DM, Jordan RL, Meltzer DM, Merritt JA, et al. Enhancement of monocyte class I and II histocompatibility antigen expression in man by in vivo interferon-s, Clinical and Experimental Immunology 69: 107115, 1987 Sriram S, Steinman L. Anti I-A ant ibody suppresses active encephalomyelitis: treatment model for diseases linked to IR genes. Journal of Experimental Medicine 158: 13621367, 1983 Steiniger B, van der Meide PH. Rat ependyma and microglial cells express class II MHC antigens after intravenous infusion of recombinant gamma interferon . Journal of Neuroimmunology 19: 111-118, 1988 Steinman LS, Rosenbaum JT , Sriram S, McDevitt HO, In vivo effects of antibodies to immune response gene products: prevention of experimental allergic encephalomyelitis. Proceedings of the National Academy of Sciences USA 78: 7111-7114, 1981 Suter CC, Westmoreland BF, Sharbrough FW, Hermann RC. Electroencephalographic abnormalities in interferon encephalopathy: a preliminary report . Mayo Clinic Proceedings 59: 847-850, 1984 Sztein MB, Steeg PS, Johnson HM, Oppenheim 11. Regulation of human peripheral blood monocyte DR antigen expression in vitro by Iymphokines and recombinant interferons . Journal of Clinical Investigation 73: 556-565, 1984 Thompson AJ, Kermode AG, Wicks D, MacManus DG, Kendall BE, et al. Major differences in the dynamics of primary and secondary progressive multiple sclerosis. Annals of Neurology 29: 53-62, 1991 Traugott U. Multiple sclerosis: relevance of class I and class II MHC-expressing cells to lesion development. Journal of Neuroimmunology 16: 283-302, 1987 Traugott U, Lebon P. Demonstration of a ,{3 and -y interferon

Drugs 44 (6) /992

in active chronic multiple sclerosis lesions. Annals of the New York Academy of Sciences 540: 309-311, 1988a Traugott U, Lebon P. Multiple sclerosis: involvement of interferons in lesion pathogenesis. Annals of Neurology 24: 243-251, 1988b Trotter JL, Clifford DB, Anderson CB, van der Veen RC, Hicks BC, et al. Elevated serum interleukin-2 levels in chronic progressive multiple sclerosis. New England Journal of Medicine 318: 1206, 1988 Tuohy VK, Lu Z, Sobel RA, Laursen RA, Lees MB. A synthetic peptide from myelin proteolipid protein induces experimental allergic encephalomyelitis. Journal of Immunology 141: 1126-1130, 1988 Vanguri P, Farber J. Identification of CRG-2: an interferoninducible mRNA predicted to encode a murine cytokine. Journal of Biological Chemistry 265: 15049-15057, 1990 Vartdal F. HLA associations in multiple sclerosis: implications for immunopathogenesis. Research in Immunology 140: 192-196, 1989 Vervliet G, Carton H, Meulepas E, Billiau A. Interferon production by cultured peripheral leucocytes of MS patients. Clinical and Experimental Immunology 58: 116-126, 1984 Vervliet G, Claeys H, Van Haver H, Carton H, Vermylen C, et al. Interferon production and natural killer (NK) cell activit y in leucocyte cultures from multiple sclerosis patients. Journal of the Neurological Sciences 60: 137·150, 1983 Vido vic M, Sparacio SM, Elovitz M, Benveniste EN. Induction and regulation of class II major histocompatibility complex mRNA expression in astrocytes by interferonand tumor necrosis factor-a . Journal of Neuro immunology 30: 189-200, 1990 Volz MA, Kirkpatrick CH. Interferons 1992. How much of the promise has been realised? Drugs 43: 285-294, 1992 Voorthuis JAC, Uitdehaag BMJ, De Groot CJA, Goede PH , Van der Meide PH, et al. Suppression of experimental allergic encephalom yelitis by intraventricular administration of interferon-gamma in Lewis rats. Clinical and Experimental Immunology 81: 183-188, 1990 Waksman B. Mechanisms in multiple sclerosis. Nature 318: 104-105, 1985 Weinshenker BG, Bass B, Rice GPA, Noseworthy J, Carriere W, et al. The natural history of multiple sclerosis: a geographically based study. 2. Predictive value of the early clinical course. Brain 112: 1419-1428, 1989 Wucherpfennig KW, Ota K, Endo N, Seidman JG , Rosenzweig A, et al. Shared human T cell receptor V{3 usage to immunodominant regions of myelin basic protein . Science 248: 1016-1019, 1990 Correspondence and reprints: Dr Hillel S. Panitch, Department of Neurology, University of Maryland School of Medicine, 22 South Greene Street, Baltimore, MD 21201, USA.