`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`The New England Journal of Medicine
`
`Review Article
`
`Drug Therapy
`
`A
`
`L A S T A I R
`
` J.J. W
`
`O O D
`
`, M.D.,
`
` Editor
`
`M
`
` M
`
`OF
`ANAGEMENT
`S
`CLEROSIS
`
`ULTIPLE
`
`
`
` A. C
`, M.D., J
` A. R
`R
`OHEN
`EFFREY
`UDICK
`ICHARD
` W
`-G
`, M.D.,
`B
`IANCA
`EINSTOCK
`UTTMAN
`R
` P. K
`, M.D.,
`EVERE
`INKEL
` R
` M. R
`, M.D.
`ICHARD
`ANSOHOFF
`
`AND
`
`, M.D.,
`
`M
`
`ULTIPLE SCLEROSIS is a common dis-
`ease of the central nervous system affect-
`ing approximately 1 million young adults,
`mostly women, worldwide.
` It is characterized by
`1
`episodic neurologic symptoms that are often fol-
`lowed by fixed neurologic deficits, increasing disabil-
`ity, and medical, socioeconomic, and physical decline
`over a period of 30 to 40 years.
`For most of the 20th century, multiple sclerosis
`was considered untreatable. In 1982, the Multiple
`Sclerosis Society of Canada and the National Multiple
`Sclerosis Society of the United States sponsored the
`first international workshop on therapeutic trials.
`2
`This workshop served to usher in an era of activism
`and optimism that has substantially replaced wide-
`spread therapeutic nihilism and skepticism about the
`feasibility of clinical trials in multiple sclerosis.
`There have been a number of important advances
`since the international workshop.
` The Expanded
`3,4
`Disability Status Scale achieved widespread use as a
`single measure of the severity of multiple sclerosis.
`5
`Magnetic resonance imaging (MRI) was invented,
` and quickly established
`applied to multiple sclerosis,
`6
`as a sensitive marker of the pathologic process. Large
`multicenter clinical trials were completed,
` and
`7-14
` The
`monographs on clinical trials were published.
`15,16
`Food and Drug Administration (FDA) approved in-
`terferon beta-1a (Avonex, Biogen, Cambridge, Mass.),
`interferon beta-1b (Betaseron, Berlex Laboratories,
`Richmond, Calif.), and glatiramer acetate (Copax-
`one, Teva Marion Partners, Kansas City, Mo.) for pa-
`
`From the Mellen Center for Multiple Sclerosis Treatment and Research,
`Department of Neurology, Cleveland Clinic Foundation, Cleveland, OH
`44106, where reprint requests should be addressed to Dr. Rudick.
`©1997, Massachusetts Medical Society.
`
`1604
`
`ⴢ
`
`November 27, 1997
`
`tients with relapsing–remitting multiple sclerosis. As
`a result of these advances, effective therapies are now
`available, and clinical trials of other promising thera-
`pies are under way.
`
`DISEASE CHARACTERISTICS
`RELATED TO TREATMENT DECISIONS
`The goal of therapy in patients with multiple scle-
`rosis is to prevent relapses and progressive worsening
`of the disease. Spontaneous recovery is rare when
`neurologic deficits have persisted for longer than six
`months, and there are no known therapies that pro-
`mote regeneration and reverse fixed neurologic def-
`icits. Therefore, disease-modifying therapy should
`be considered before neurologic deficits have per-
`sisted longer than six months. Decisions in individ-
`ual patients should be based both on the course of
`the patient’s disease and on the probability of severe
`disabling disease.
`A standardized nomenclature to describe the
`course of multiple sclerosis (Table 1) was developed
`by consensus.
` The most common pattern at onset
`17
`is relapsing–remitting disease, but it becomes sec-
`ondary progressive disease over time in more than 50
`percent of patients. Approximately 10 percent of pa-
`tients have primary progressive multiple sclerosis.
`They tend to be older at onset (40 to 60 years of age)
`and commonly have a progressive myelopathy. Pa-
`tients with primary progressive multiple sclerosis
`have fewer gadolinium-enhanced lesions on cranial
`MRI scans and fewer inflammatory changes in cere-
`brospinal fluid than patients with secondary progres-
`sive multiple sclerosis.
` Progressive relapsing multi-
`18
`ple sclerosis is a very uncommon pattern of disease.
`The vast majority of patients have relapsing–remit-
`ting multiple sclerosis during the early years and sec-
`ondary progressive multiple sclerosis later. Patients
`with relapsing–remitting multiple sclerosis have the
`best responses to treatment, whereas patients with
`progressive disease are less responsive to treatment.
`Disease-modifying therapy should be considered
`early in the course for patients with an unfavorable
`prognosis. The unfavorable prognostic markers relat-
`ed to more rapid worsening of disease that are listed
`in Table 2 can be used to select patients for treat-
`ment.
` Patients who have multiple cranial MRI le-
`19-22
`sions at the time of their first symptoms are much
`more likely to have major disability later on.
` There-
`23
`fore, in addition to the clinical features, the findings
`on cranial MRI are useful in selecting patients for ear-
`ly treatment. Approximately 10 percent of patients
`have relatively benign disease, however, so not every
`patient should receive disease-modifying therapy.
`
`Hopewell EX1017
`
`1
`
`

`

`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`DRUG THERAPY
`
`DESIGN OF CLINICAL TRIALS
`
`Relation between Treatment Strategies and Pathogenesis
`Various lines of research support the hypothesis
`that multiple sclerosis is an autoimmune process oc-
`curring in genetically susceptible persons after an en-
`vironmental exposure. The geographic heterogeneity
`of the disease and the widely varying prevalence rates
`in different ethnic populations suggest interplay be-
`tween environmental and genetic factors. The obser-
`vation that common viral infections can precipitate
`relapses led to the concept that viruses can trigger
`autoimmune demyelination in susceptible persons.
`24
`No specific pathogen has been reliably linked to mul-
`tiple sclerosis, however, so none of the current treat-
`ment approaches are targeted at microbial patho-
`gens. The children of patients with multiple sclerosis
`have an increased risk (30-fold to 50-fold) of multi-
` Studies of twins and adopted children
`ple sclerosis.
`25
`suggest that the increased risk is largely genetic.
`26
`Candidate-gene and whole-genome screening sug-
`gests that multiple weakly acting genes interact to
`determine the risk of multiple sclerosis.
`27-29
`Most current therapeutic approaches are based on
`the hypothesis that multiple sclerosis is an organ-
`specific autoimmune disease. Inoculation of suscep-
`tible animals with myelin proteins results in a relaps-
`ing–remitting, inflammatory, demyelinating central
`nervous system disease called experimental autoim-
`mune encephalomyelitis. Experimental autoimmune
`encephalomyelitis can be transferred to unimmu-
`nized animals through activated T cells that recog-
`nize small fragments of myelin proteins.
` These
`30
`pathogenic T cells use a restricted array of genes for
`T-cell–antigen receptors. Molecular strategies that in-
`terrupt the interaction between myelin protein pep-
`tides and pathogenic T-cell receptors are effective in
`acute experimental autoimmune encephalomyelitis.
`31
`Human T cells that recognize myelin antigens also
`have restricted use of T-cell receptors, prompting at-
`tempts to eliminate pathogenic T cells with antibod-
` To date, however, evidence of a
`ies or vaccination.
`32
`unique immunologic abnormality in patients with
`multiple sclerosis is lacking. In particular, T cells that
`recognize myelin can be isolated with similar fre-
`quencies from patients with multiple sclerosis and
`normal subjects. Furthermore, as autoimmune dis-
`eases progress, self-peptides are released from the
`target organ, increasing the diversity of the T-cell re-
`sponse. This phenomenon, termed “epitope spread-
`ing,” occurs in animals with chronic experimental
`autoimmune encephalomyelitis
` and in patients
`33,34
`with multiple sclerosis
` and may limit simple treat-
`35
`ment strategies based on blocking the recognition
`of autoantigen.
`The lesions of multiple sclerosis resemble those in-
`duced by delayed hypersensitivity, containing inflam-
`matory cytokines, activated T cells, and mononuclear
`
` C
`T
` 1.
`ABLE
`
` C
`
`LINICAL
`ATEGORIES
`
`OF
`
` M
` S
`.
`ULTIPLE
`CLEROSIS
`
`D
`ISEASE
`
` C
`ATEGORY
`
`Relapsing–remitting
`
`Secondary progressive
`
`Primary progressive
`
`Progressive relapsing
`
`D
`EFINITION
`
`Episodes of acute worsening with recovery and
`a stable course between relapses
`Gradual neurologic deterioration with or with-
`out superimposed acute relapses in a patient
`who previously had relapsing–remitting mul-
`tiple sclerosis
`Gradual, nearly continuous neurologic deterio-
`ration from the onset of symptoms
`Gradual neurologic deterioration from the on-
`set of symptoms but with subsequent super-
`imposed relapses
`
`T
`
` P
` T
` P
` M
`
` 2.
`REDICT
`HAT
`ROGNOSTIC
`ARKERS
`ABLE
`M
` S
` M
` S
`.
`CLEROSIS
`ORE
`EVERE
`ULTIPLE
`
`Progressive disease from the onset of symptoms
`Motor and cerebellar signs at presentation to neurologist
`Short interval between the first two relapses
`Poor recovery from relapse
`-weighted MRI at presentation
`Multiple cranial lesions on T
`2
`
`phagocytes.
` These elements, shown in Figure 1,
`36,37
`are all potential targets for intervention. Function-
`related T-cell surface molecules can be down-regulat-
`ed with antibodies. Cytokine-based therapies, such as
`those involving soluble receptors for tumor necrosis
`factor
` or immunosuppressive cytokines such as
`a
`transforming growth factor
` or interleukin-10, may
`b
`potentially be effective. The inflammation of the cen-
`tral nervous system may also be sensitive to interven-
`tion directed against leukocyte and cerebrovascular
`endothelial adhesion molecules or chemokines, which
`mediate the migration of leukocytes into the central
`nervous system. In both multiple sclerosis and exper-
`imental autoimmune encephalomyelitis, myelin anti-
`bodies are concentrated in the central nervous sys-
`tem, and demyelinating antibodies in experimental
`autoimmune encephalomyelitis synergize with T-cell–
` Pathogenic antibodies are also
`effector mechanisms.
`47
`potential therapeutic targets.
`
`Controlled Clinical Trials
`Therapeutic advances in multiple sclerosis are de-
`pendent on clinical trials because of the highly vari-
`able and unpredictable course of the disease and the
`difficulty in precisely measuring neurologic disability.
`The Expanded Disability Status Scale,
` the most wide-
`5
`ly used outcome measure in clinical trials of multiple
`sclerosis, is an ordinal rating scale ranging from 0 to
`10, in increments of 0.5, with higher scores reflecting
`increasing severity. The fact that there is a single
`score for each patient at each time point makes study
`design and statistical analysis relatively simple, but
`
`Volume 337 Number 22
`
`ⴢ
`
`1605
`
`2
`
`

`

`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`The New England Journal of Medicine
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`Systemic
`Circulation
`
`Blood–Brain
`Barrier
`
`Central Nervous
`System
`
`Recruitment of cells
`
`Myelin injury
`
`Chemokines
`
`T cell
`
`Activation by
`superantigens, molecular
`mimicry, or unknown
`mechanisms
`
`Astrocyte
`
`Microglia
`
`Activation of
`endothelium
`
`Cytokines
`
`4
`
`1
`
`2
`
`1
`
`Conduction block
`
`3
`
`Activated
`autoreactive T cell
`
`Autoreactive
`T cell
`
`Antigen-presenting cell
`
`Figure 1.
` Pathogenesis of Multiple Sclerosis.
` or unknown mechanisms.
` molecular mimicry,
`Circulating autoreactive T cells are activated by stimulation with superantigens,
`39
`38
`Once activated, these autoreactive cells traverse the blood–brain barrier to enter the central nervous system. Perivascular antigen-
`presenting cells provide the signals necessary to result in the activation and clonal expansion of these autoreactive T cells and the
`secretion of proinflammatory cytokines by them. The cytokines, including tumor necrosis factor and interferon-
`, induce astrocytes
`g
`and leukocytes to secrete chemokines
` and stimulate the expression of adhesion molecules by endothelial cells. Activated micro-
`40
`glia and macrophages damage myelin internodes.
` Proinflammatory cytokines may directly inhibit nerve conduction, leading to
`41
`neurologic dysfunction. Immunosuppressive cytokines (not shown) inhibit the inflammatory process, leading to neurologic recov-
`ery. The putative mechanisms of action of the therapeutic effects of interferon beta, as indicated by the numbers, include inhibition
`of the proliferation of autoreactive T cells (1)
`; inhibition of the expression of major-histocompatibility-complex class II molecules,
`42
`43
`leading to reduced antigen presentation within the central nervous system (2); inhibition of metalloproteases,
` leading to re-
`44,45
`duced migration of T cells into and through the central nervous system (3); and induction of immunosuppressive cytokines,
` lead-
`46
`ing to resolution of the inflammatory process (4).
`
`the minimal changes in scores for some patients over
`long intervals and the subjectivity in making the clin-
`ical ratings limit the value of the scale. The usefulness
`of the Expanded Disability Status Scale has been im-
`proved by the addition of a definition of treatment
`failure as sustained worsening of a clinically impor-
` An effort is under way to develop im-
`tant amount.
`48
`proved clinical outcome measures,
` which could de-
`49
`crease the required sample sizes or shorten the
`duration of multiple sclerosis trials.
`Serial MRI studies have shown that new gadolin-
`ium-enhanced lesions are 5 to 10 times as common
`
`as clinical relapses.
` Preliminary evidence of the ef-
`50
`ficacy of treatments on the basis of MRI findings
`will probably serve as the basis for future trials, and
`all will include serial MRI as an important secondary
`outcome measure.
`
`RELAPSING MULTIPLE SCLEROSIS
`
`Corticosteroids
`Corticosteroids are the mainstay of treatment for
`acute relapses of multiple sclerosis. Corticosteroids
`have immunomodulatory and antiinflammatory ef-
`
`1606
`
`ⴢ
`
`November 27, 1997
`
`3
`
`

`

`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`DRUG THERAPY
`
`fects that restore the blood–brain barrier, reduce
`edema, and may possibly improve axonal conduc-
`tion. Corticosteroid therapy shortens the duration
`of the relapse and accelerates recovery, but whether
`the overall degree of recovery is improved or the
`long-term course is altered is not known.
` Corti-
`51-53
`cotropin was demonstrated to help recovery from re-
` but it has been largely replaced by high-dose
`lapse,
`52
`intravenous methylprednisolone, because the latter
`has a more rapid onset of action, produces more
`consistent benefits, and has fewer side effects.
`52,54
`For moderate-to-severe relapses, 1000 mg of meth-
`ylprednisolone per day by intravenous infusion for
`3 to 5 days followed by 60 mg of oral prednisone
`per day, with tapering of the dose over a period of
`12 days, accelerates neurologic recovery.
`In the Optic Neuritis Treatment Trial, 457 patients
`with acute optic neuritis were randomly assigned to
`receive 1000 mg of intravenous methylprednisolone
`per day for 3 days followed by 1 mg of oral predni-
`sone per kilogram of body weight per day for 11
`days; 1 mg of oral prednisone per kilogram per day
`for 14 days; or oral placebo. The rate of recovery of
`vision was significantly faster in the intravenous-
`methylprednisolone group, with the greatest bene-
`fits in patients with visual acuity of 20/50 or worse
` but there were no significant differences
`at entry,
`9
`between groups in visual outcome at six months.
`Prednisone therapy increased the risk of new epi-
`sodes of optic neuritis in either eye, and intravenous
`methylprednisolone reduced by approximately 50
`percent the risk of an attack leading to the diagnosis
`of multiple sclerosis during the two-year follow-up.
`55
`This effect was most evident in patients at highest
`risk for subsequent relapse — those with multicen-
`tric brain lesions on MRI at entry into the study. Af-
`ter three years, differences between the treatment
`groups were no longer significant,
` suggesting that
`56
`intravenous methylprednisolone delayed but did not
`stop the development of multiple sclerosis after op-
`tic neuritis. These results have led to the widespread
`use of intravenous methylprednisolone for patients
`with optic neuritis and abnormal findings on MRI
`of the brain. The results also renewed debate over
`whether intravenous methylprednisolone has long-
`term benefits for patients with multiple sclerosis. A
`clinical trial is under way to determine whether
`pulsed doses of intravenous methylprednisolone giv-
`en every other month slow disease progression in
`patients with moderate disability and secondary pro-
`gressive multiple sclerosis.
`
`Interferon Beta
`Interferon beta is the treatment of choice for pa-
`tients with relapsing–remitting multiple sclerosis.
`Two forms of recombinant interferon beta — 1a and
`1b — have been approved by the FDA and Europe-
`an regulatory agencies. Interferon beta-1a is a glyco-
`
`sylated, recombinant mammalian-cell product, with
`an amino-acid sequence identical to that of natural
`interferon beta. Interferon beta-1b is a nonglycosy-
`lated recombinant bacterial-cell product in which
`serine is substituted for cysteine at position 17.
`Interferon beta-1b was tested in a multicenter trial
`involving 372 patients with relapsing–remitting mul-
`tiple sclerosis and mild-to-moderate disability. Treat-
`ment consisted of either 8 million units (250
`g) or
`m
`1.6 million units (50
`g) of interferon beta-1b or pla-
`m
`cebo given by subcutaneous injection every other day
`for up to five years. As compared with treatment with
`placebo, treatment with the higher dose reduced the
`relapse rate by 31 percent, increased the proportion
`of patients who were relapse-free (27 percent vs. 17
`percent), and reduced by a factor of 2 the number of
`patients who had moderate and severe relapses.
`8
`There was no difference in the proportion of patients
`in whom disability increased or in changes in the dis-
`ability scores between treatment groups. The patients
`in the placebo group had a mean increase of 17 per-
`-weighted MRI at
`cent in the area of the lesions on T
`2
`three years, as compared with a mean decrease of
`6 percent in the patients given high-dose interferon
`beta-1b. There was also a significant reduction in dis-
`ease activity, defined as the finding of new or enlarg-
`ing lesions in serial MRIs.
` The MRI findings in this
`57
`study were pivotal in obtaining FDA approval for in-
`terferon beta-1b and initiated the era in which MRI
`has a key role in assessing therapeutic responses in pa-
`tients with multiple sclerosis.
`Interferon beta-1a was tested in a multicenter trial
`involving 301 patients with relapsing–remitting mul-
`tiple sclerosis and mild-to-moderate disability. Treat-
`ment consisted of weekly intramuscular injections
`(6 million units [30
`g]) or placebo for up to two
`m
`years.
` The principal outcome was the length of
`7,58
`time to the progression of disability, defined as a de-
`crease from base line of at least 1.0 point on the Ex-
`panded Disability Status Scale that persisted for at
`least six months. Treatment with interferon beta-1a,
`as compared with placebo, significantly lowered the
`probability of progression of disability
` and of severe
`7
` In addition, patients treated with inter-
`disability.
`59
`feron beta-1a for two years had a reduction of 32
`percent in the annual rate of relapse, and had fewer
`gadolinium-enhanced lesions on MRI. The favor-
`able effect of interferon beta-1a on gadolinium-
`enhanced lesions, confirmed in a separate study with
`interferon beta-1b,
` suggests that interferon beta
`60
`inhibits new lesion formation.
`Both types of interferon beta are usually well tol-
`erated. The most common side effects are influenza-
`like symptoms for 24 to 48 hours after each injec-
`tion, and these usually subside after two to three
`months of treatment. Injection of interferon beta-1b
`causes redness, tenderness, swelling, and occasional-
`ly, necrosis at the injection site. Interferon beta-1b
`
`Volume 337 Number 22
`
`ⴢ
`
`1607
`
`4
`
`

`

`The New England Journal of Medicine
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`TABLE 3. IMPORTANT UNRESOLVED QUESTIONS
`RELATED TO INTERFERON BETA THERAPY IN PATIENTS
`WITH RELAPSING–REMITTING MULTIPLE SCLEROSIS.
`
`When should therapy be started?
`How long should therapy be continued?
`Can the dose be individualized to achieve maximal therapeu-
`tic benefit?
`What are the therapeutic mechanisms of action of the drug?
`What are the long-term benefits?
`Which preparation of interferon beta is clinically superior?
`
`can also cause slight elevations in serum aminotrans-
`ferase concentrations, leukopenia, or anemia, and a
`few patients have become depressed or have had
`worsening of preexisting depression. Interferon beta-
`1b–neutralizing activity was detected in serum sam-
`ples from 38 percent of patients by the third year of
`treatment
` and correlated with decreased efficacy of
`61
`therapy.
` Serum interferon beta-1a–neutralizing ac-
`62
`tivity was found less often — in 14 percent of pa-
`tients after one year and 22 percent after two years.
`7
`Although interferon beta therapy is effective, im-
`portant questions remain (Table 3). A risk–benefit
`analysis must be done in each patient. The cost of
`therapy, currently approximately $8,000 to $10,000
`per year, and the uncertain long-term risks may out-
`weigh the benefits in patients with mild multiple scle-
`rosis and a favorable prognosis. Whether long-term
`therapy should be started at the time of the first at-
`tack and what constitutes the optimal duration of
`therapy are not known. In a study of recombinant
`interferon alfa-2a (Roferon-A, Hoffmann–LaRoche,
`Nutley, N.J.) in patients with relapsing–remitting
`multiple sclerosis, relapse occurred when therapy was
` suggesting the need for
`stopped after six months,
`63
`more prolonged therapy. The final report from the in-
`terferon beta-1b study
` suggested that patients con-
`61
`tinued to respond to treatment for five years, a finding
`that supports the value of long-term therapy, but the
`high dropout rate (greater than 50 percent) may
`have biased the results in favor of long-term therapy.
`Specific indications to stop therapy were steady pro-
`gression of disability over a period of six months or
`treatment with three courses of corticotropin or cor-
`ticosteroids for acute relapses during a one-year pe-
` The appearance of serum interferon beta–neu-
`riod.
`64
`tralizing antibodies should prompt alternative therapy,
`particularly in patients with disease progression.
`65
`The variable biologic response to interferon beta
`suggests that the dose could be individualized. Side
`effects of the interferon beta-1b correlate with body-
` but there are no established methods to
`surface area,
`61
`individualize the dose for maximal efficacy. Figure 1
`shows the putative sites of action of interferon beta in
`patients with multiple sclerosis. Clarifying the mecha-
`nisms most closely linked to efficacy might lead to
`
`1608
`
`ⴢ
`
`November 27, 1997
`
`better methods to individualize treatment, particularly
`if the therapeutic effect could be monitored easily.
`The best preparation of interferon beta and the
`long-term benefits of such therapy remain controver-
`sial. Both interferon beta-1a and interferon beta-1b
`reduce the relapse rate and disease activity on MRI,
`but interferon beta-1a appears to be better tolerated.
`In addition, interferon beta-1a results in less progres-
`sion of disability,
` suggesting that long-term therapy
`7
`will lessen the eventual impact of the disease.
`
`Glatiramer Acetate
`Glatiramer acetate is a mixture of random syn-
`thetic polypeptides composed of
`-alanine,
`-glutam-
`L
`L
`ic acid, L-lysine, and L-tyrosine in a molar ratio of
`6.0:1.9:4.7:1.0. It was synthesized as an immuno-
`chemical mimic of myelin basic protein, a putative
`autoantigen in multiple sclerosis. After glatiramer
`acetate was found to inhibit experimental autoim-
`mune encephalomyelitis, a small trial suggested effi-
`cacy in patients with relapsing–remitting multiple
`sclerosis.66 It was subsequently tested in a trial involv-
`ing 251 patients with relapsing–remitting multiple
`sclerosis and mild-to-moderate disability. Treatment
`consisted of daily subcutaneous injections of 20 mg
`of glatiramer acetate or placebo for two years.14 The
`annualized relapse rate, the primary end point, was
`29 percent lower in the glatiramer acetate group,
`and the proportion of patients who did not have a
`relapse was higher (34 percent vs. 27 percent). A
`greater proportion of patients in the glatiramer ace-
`tate group had an improvement of 1.0 point or more
`in their score on the Expanded Disability Status Scale
`(25 percent vs. 15 percent), and fewer had worsening
`of disability (21 percent vs. 29 percent). The most
`common side effect was mild reactions at the injec-
`tion site, which occurred in 90 percent of patients
`given glatiramer acetate; 15 percent had brief epi-
`sodes of flushing, chest tightness, shortness of breath,
`palpitations, and anxiety after one or more injec-
`tions. Serum antibodies to glatiramer acetate also
`developed, but the presence of these antibodies had
`no effect on the clinical benefit. MRI scans, which
`were obtained at only one of the study sites, showed
`little change over the course of the study.67
`Glatiramer acetate was approved by the FDA in
`1996. It represents an alternative to interferon beta
`therapy for patients with relapsing–remitting multi-
`ple sclerosis and may be most useful for patients who
`become resistant to interferon beta treatment owing
`to serum interferon beta–neutralizing activity.
`
`Azathioprine
`Azathioprine, a purine analogue, depresses both
`cell-mediated and humoral immunity. A meta-analy-
`sis of five randomized, double-blind, placebo-con-
`trolled trials supported the conclusion that oral aza-
`thioprine (2 to 3 mg per kilogram per day) reduces
`
`5
`
`

`

`DRUG THERAPY
`
`the rate of relapse in multiple sclerosis 68 but has no
`effect on the progression of disability. The concern
`that prolonged azathioprine therapy may increase the
`risk of non-Hodgkin’s lymphoma or skin cancer 69
`was not confirmed in a case–control study.70 Azathi-
`oprine should be considered in patients with relaps-
`ing–remitting multiple sclerosis who do not respond
`to therapy with interferon beta or glatiramer acetate.
`Azathioprine may be useful in patients with Devic’s
`disease,71 a variant of multiple sclerosis affecting the
`optic nerves and cervical spinal cord, or in those with
`recurrent inflammatory myelitis.72
`
`Intravenous Immune Globulin
`Intravenous immune globulin has been used suc-
`cessfully in neuroimmunologic disorders, including
`acute and chronic inflammatory demyelinating poly-
`radiculopathy and myasthenia gravis, but its role in
`patients with multiple sclerosis is not yet clear. In a
`study of 150 patients with relapsing–remitting mul-
`tiple sclerosis who were treated with intravenous im-
`mune globulin (150 to 200 mg per kilogram per
`month) or placebo for two years, there was less wor-
`sening in the scores on the Expanded Disability Sta-
`tus Scale in the group given immune globulin, but
`the differences between groups were small. The
`number of relapses and the annual relapse rate were
`also lower in the immune globulin group. However,
`the treating physicians were aware of the treatment
`assignment, which could have biased the results.
`
`PROGRESSIVE MULTIPLE SCLEROSIS
`Investigators have only recently made a clinical
`distinction between secondary progressive multiple
`sclerosis and primary progressive multiple sclerosis,
`and most studies have not distinguished between
`these two forms of chronic progressive multiple scle-
`rosis. Treatment for chronic progressive multiple
`sclerosis has usually consisted of nonspecific immune
`suppression and has been of only moderate benefit.
`
`Methotrexate
`Methotrexate inhibits dihydrofolate reductase. Low-
`dose oral methotrexate is relatively nontoxic and ef-
`fective in rheumatoid arthritis and psoriasis, presum-
`ably by inhibiting both cell-mediated and humoral
`immunity or as a result of its antiinflammatory effects.
`Sixty ambulatory patients with chronic progressive
`multiple sclerosis and moderate-to-severe disability
`were treated with methotrexate (7.5 mg weekly) or
`placebo for two years.73 As compared with placebo,
`methotrexate significantly reduced sustained worsen-
`ing, according to a composite measure of outcome
`that included the Expanded Disability Status Scale,
`the Ambulation Index,74 and two tests of arm func-
`tion. The methotrexate-treated patients had less dis-
`ease progression, and patients with secondary pro-
`gressive multiple sclerosis benefited most.73,75 The
`
`clinical benefits were considered moderate, but
`toxicity was minimal. Because there are no nontoxic
`alternative treatments for patients with chronic
`progressive multiple sclerosis, low-dose oral metho-
`trexate should be considered for those with progres-
`sive deterioration.
`
`Cyclophosphamide
`Cyclophosphamide is an alkylating agent with po-
`tent cytotoxic and immunosuppressive effects. In
`several controlled but unblinded trials, high-dose in-
`travenous cyclophosphamide with or without subse-
`quent booster injections was effective in patients
`with chronic progressive multiple sclerosis.74,76,77 How-
`ever, one trial failed to demonstrate benefit.10 In this
`trial, the examining neurologist was unaware of the
`treatment assignments, and the patients wore bath-
`ing caps to mask the alopecia induced by cyclophos-
`phamide. Cyclophosphamide has many side effects
`in addition to alopecia, including nausea and vomit-
`ing, hemorrhagic cystitis, leukopenia, myocarditis,
`infertility, and pulmonary interstitial fibrosis. Treat-
`ment with cyclophosphamide may be most appro-
`priate for patients with rapidly progressive disease
`who do not respond to less toxic alternatives such as
`methotrexate.78
`
`Cyclosporine
`Cyclosporine is a potent immunosuppressive drug
`that inhibits several steps in the activation of T cells.
`In a multicenter U.S. study, 550 patients with chronic
`progressive multiple sclerosis were given oral cyclo-
`sporine (initial dose, 7 mg per kilogram per day) or
`placebo for two years.11 The length of time before
`patients became dependent on a wheelchair was
`slightly longer in the cyclosporine group, but neph-
`rotoxicity occurred in 84 percent of cyclosporine-
`treated patients, and hypertension was also frequent.
`The moderate clinical benefits of cyclosporine ap-
`pear to be outweighed by its toxicity.
`
`FUTURE DIRECTIONS
`Long-term placebo-controlled trials may no long-
`er be ethically justifiable in patients with relapsing–
`remitting multiple sclerosis, given the evidence of
`the efficacy of interferon beta and glatiramer acetate.
`Comparisons between treatments are needed but
`will require very large samples, lengthy studies, and
`more sensitive outcome measures.49,79 Analyses of
`cost effectiveness and the quality of life80,81 will be
`increasingly important as more therapeutic options
`become available.
`There are many potential new therapies, and it is
`not clear how best to screen them. Effectiveness in
`experimental autoimmune encephalomyelitis has been
`considered a prerequisite, but many therapies that
`are effective in experimental autoimmune encepha-
`lomyelitis are not useful in patients with multiple
`
`Volume 337 Number 22
`
`ⴢ 1609
`
`6
`
`

`

`The New England Journal of Medicine
`
`sclerosis. Using serial gadolinium-enhanced MRI
`scans to screen therapies for efficacy has been rec-
`ommended,79 but such an approach has the poten-
`tial of screening out treatments that might alter a
`different stage of the pathologic process.
`Specific forms of immune suppression are based
`on deleting or inhibiting autoreactive pathogenic
`T cells. This approach may be limited by the heter-
`ogeneity and complexity of myelin determinant rec-
`ognition in patients with established multiple scle-
`rosis.35 Nevertheless, ongoing trials are testing the
`efficacy of a T-cell–receptor vaccine,32,82 myelin pep-
`tides conjugated to major-histocompatibility-com-
`plex molecules,83 and oral myelin,84 all of which are
`thought to inhibit myelin-reactive T cells. Studies
`of nonspecific immunomodulation with linomide,
`transforming growth factor b, and soluble tumor ne-
`crosis factor–receptor are under way and are planned
`for interleukin-10. Remyelination strategies are large-
`ly at a preclinical stage, although the efficacy of intra-
`venous immune globulin is being tested in patients
`who have fixed neurologic deficits.85 In the future,
`glial-cell transplantation86 or treatment with recom-
`binant growth factors87 could be used to stimulate re-
`generation and functional recovery.
`
`We are indebted to Drs. Henry F. McFarland and W. Ian Mc-
`Donald for their thoughtful review of the manuscript and many
`helpful suggestions.
`
`REFERENCES
`
`1. Williams R, R