52'! "-93
`
`Neuralwy
`,<-IE§‘ML Cur" Yti'ai"
`‘ UC: San Diego
`Rec-eiyed on: 04-
`
`Flcam‘.
`
`) Iciat Journal’of the American Academy of Neurology' a
`ff
`i
`i;
`,
`
`i
`
`?
`
`Apotex v. Novartis
`lPR2017-00854
`
`NOVARTIS 2073
`
`Apotex v. Novartis
`IPR2017-00854
`NOVARTIS 2073
`
`1
`
`

`

`1 'eurology®
`contents
`
`APRIL 1993
`
`OFFICIAL JOURNAL OF THE AMERICAN ACADEMY OF NEUROLOGY
`
`VOLUME 43 • NUMBER 4
`
`641
`
`644
`
`Editorial
`Interferon beta in multiple sclerosis
`B.G. W. Amason
`
`Views & Reviews
`Clinical Neuromythology X. Faithful fashion:
`Survival status of the brain transplant cure for
`parkinsonism W.M. Landau
`
`Issues of Neurological Practice
`650 Medicare hospital utilization review for ischemic
`cerebrovascular disease D.J. Lanska
`
`655
`
`662
`
`Articles
`Interferon beta-lb is effective in relapsing(cid:173)
`remitting multiple sclerosis. I. Clinical results of
`a multicenter, randomized, double-blind, placebo(cid:173)
`controlled trial The IFNB Multiple Sclerosis Study
`Group
`
`Interferon beta-lb is effective in relapsing(cid:173)
`remitting multiple sclerosis. II. MRI analysis
`results of a multicenter, randomized, double(cid:173)
`blind, placebo-controlled trial D. W. Paty, D.K.B. Li,
`the UBC MS I MRI Study Group, and the IFNB Multiple
`Sclerosis Study Group
`
`668 Hyperkalemic periodic paralysis: Rapid molecular
`diagnosis and relationship of genotype to pheno(cid:173)
`type in 12 families W.G. Feero, J. Wang, F. Barany,
`J. Zhou, S .M. Todorovic, R. Conwit, G. Galloway,
`I. Hausmanowa-Petrusewicz, A Fidzianska, K. Arahata,
`H.B. Wessel, C. Wadelius, H.G. Marks, P. Hartlage,
`H. Hayakawa, and E.P. Hoffman
`
`674
`
`677
`
`Levodopa, melanoma, and Parkinson's disease
`W.J. Weiner, C. Singer, J.R. Sanchez-Ramos,
`and J.N. Goldenberg
`
`Clinical experience with controlled-release
`carbidopa/levodopa in Parkinson's disease
`R. Pahwa, K. Busenbark, S.J. Huber, D. Michalek,
`J .P. Hubble, and W.C. Koller
`
`681 Magnetic resonance imaging in childhood
`intractable partial epilepsies: Pathologic correla(cid:173)
`tions R. Kuzniecky, A Murra, D. King, R . Morawetz,
`J. Smith, R. Powers, F. Yaghmai, E. Faught,
`B. Gallagher, and O.C. Snead
`
`688
`
`693
`
`Felbamate monotherapy for partial-onset
`seizures: An active-control trial E. Faught,
`R . C. Sachdeo, M.P. Remler, S . Chayasirisobhon,
`V.J. Iragui-Madoz, R.E. Ramsay, T.P. Sutula,
`A Kanner, R.N. Harner, R. Kuzniecky, L.D. Kramer,
`M. Kamin, and A Rosenberg
`
`Felbamate: A double-blind controlled trial in
`patients undergoing presurgical evaluation of
`partial seizures B. Bourgeois, I.E. Leppik,
`J.C. Sackellares, K. Laxer, R. Lesser, J .A. Messenheimer,
`L.D. Kramer, M. Kamin, and A Rosenberg
`
`697
`
`701
`
`T 2 relaxation time in patients with Parkinson's
`disease A Antonini, K.L. Leenders, D. Meier,
`W.H. Oertel, P. Boesiger, and M. Anliker
`
`Sleep state and the risk of seizure recurrence
`following a first unprovoked seizure in childhood
`S. Shinnar, AT. Berg, Y. Ptachewich, and M. Alemany
`
`706 Material-specific memory in the intracarotid
`amobarbital procedure K. Perrine, J . Gershengorn,
`E.R. Brown, I.S. Choi, D.J. Luciano, and 0 . Devinsky
`
`712
`
`Clinical course of spontaneous subarachnoid
`hemorrhage: A population-based study in
`King County, Washington W.T. Longstreth, Jr. ,
`L.M. Nelson, T.D. Koepsell, and G. van Belle
`
`719 MRI volumetric measurement of amygdala and
`hippocampus in temporal lobe epilepsy F. Cendes,
`F. Andermann, P. Gloor, A Evans, M. Jones-Gotman,
`C. Watson, D. Melanson, A Olivier, T. Peters,
`I. Lopes-Cendes, and G. Leroux
`
`725
`
`728
`
`733
`
`742
`
`747
`
`751
`
`755
`
`Facial asymmetry, hippocampal pathology, and
`remote symptomatic seizures: A temporal lobe .
`epileptic syndrome G.D. Cascino, R.R . Luckstein,
`F. W. Sharbrough, and C.R. Jack, Jr.
`
`Brain infarction severity differs according to
`cardiac and arterial embolic source S .G. Tim sit,
`R.L. Sacco, J.P. Mohr, M.A. Foulkes, T.K. Tatemichi,
`P.A. Wolf, T.R. Price, and D.B. Hier
`
`Selective dorsolateral frontal lobe dysfunction
`associated with diencephalic amnesia E.P. Pepin
`and L. Auray-Pepin
`
`Cerebral venous thrombosis in paroxysmal noc(cid:173)
`turnal hemoglobinuria: Report of two cases
`M. Al-Hakim, M.B. Katirji, I. Osorio, and R . Weisman
`
`Gelastic seizures and hypothalamic hamartomas:
`Evaluation of patients undergoing chronic
`intracranial EEG monitoring and outcome of
`surgical treatment G.D. Cascino, F. Andermann,
`S.F. Berkovic, R.I. Kuzniecky, F. W. Sharbrough,
`D.L. Keene, P.F. Bladin, P.J. Kelly, A. Olivier,
`and W. Feindel
`
`The natural history of amyotrophic lateral sclero(cid:173)
`sis and the use of natural history controls in
`therapeutic trials J . Pradas, L. Finison, P.L. Andres,
`B. Thornell, D. Hollander, and T.L. Munsat
`
`The fatigue of rapid repetitive movements
`R.G. Miller, R.S. Moussavi, A.T. Green, P.J. Carson,
`and M. W. Weiner
`
`762 Ventrolateral and dorsomedial somatosensory
`association cortex damage produces distinct
`somesthetic syndromes in humans R.J. Caselli
`
`771
`
`HLA-DPBl allele associates with early-onset
`myasthenia gravis in Japan T. Horiki, J . Moriuchi,
`H. Inoko, K. Morita, K. Tsuji, Y. Shinohara,
`Y. Ichikawa, and S. Arimori
`
`continued on page BA
`
`April 1993 NEUROLOGY 43 7A
`
`2
`
`

`

`Neurology contents continued
`
`775
`
`779
`
`785
`
`791
`
`795
`
`801
`
`809
`
`817
`
`825
`
`829
`
`831
`
`834
`
`Brain magnetic resonance imaging in coronary
`artery bypass grafts: A pre- and postoperative
`assessment R. Schmidt, F. Fazekas, H. Offenbacher,
`H. Mii.chler, W. Freidl, F. Payer, B. Rigler,
`M.I.G. Harrison, and H. Lechner
`
`Auditory system degeneration in Alzheimer's
`disease U.K. Sinha, K.M. Hollen, R. Rodriguez,
`and C.A. Miller
`
`Polyglucosan body disease simulating amyotro(cid:173)
`phic lateral sclerosis T.D. McDonald, P.L. Faust,
`C. Bruno, S. DiMauro, and J.E. Goldman
`
`Kennedy's disease: A clinicopathologic correlation
`with mutations in the androgen receptor gene
`A.A. Amato, T. W. Prior, R.J. Barohn, P. Snyder,
`A. Papp, and J.R. Mendell
`
`Duchenne muscular dystrophy: Deficiency of
`dystrophin-associated proteins in the sarcolemma
`K. Ohlendieck, K. Matsumura, V. V. Ionasescu,
`J .A. Towbin, E.P. Bosch, S.L. Weinstein, S. W. Sernett,
`and K.P. Campbell
`
`Altered brain development in Turner's syndrome:
`An event-related potential study R. Johnson, Jr.,
`J. W. Rohrbaugh, and J.L. Ross
`
`Localization of GMl and Gal(J31-3)GalNAc anti(cid:173)
`genic determinants in peripheral nerve M. Corbo,
`A. Quattrini, N. Latou, and A.P. Hays
`
`The prevalence by staged severity of various
`types of diabetic neuropathy, retinopathy, and
`nephropathy in a population-based cohort: The
`Rochester Diabetic Neuropathy Study P.J. Dyck,
`K.M. Kratz, J.L. Karnes, W.J . Litchy, R. Klein,
`J .M. Pach, D.M. Wilson, P.C. O'Brien,
`and L.J. Melton III
`
`Myopathy in the elderly: Evaluation of the
`histopathologic spectrum and the accuracy of
`clinical diagnosis D. Lacomis, D.A. Chad,
`and T. W. Smith
`
`Brief Communications
`Prognostic significance of the onset mode in
`parkinsonism A.H. Rajput, R. Pahwa, P. Pahwa,
`and A. Rajput
`
`Long-term safety of azathioprine therapy in
`multiple sclerosis M.P. Amato, G. Pracucci,
`G. Ponziani, G. Siracusa, L. Fratiglioni,
`and L. Amaducci
`
`Longitudinal experience with botulinum toxin
`injections for treatment of blepharospasm and
`cervical dystonia J . Jankouic and K.S. Schwartz
`
`836
`
`838
`
`842
`
`Contrast agent overdose causing brain retention
`of contrast, seizures, and parkinsonism E.F. May,
`G.S.F. Ling, C.A. Geyer, and B. Jabbari
`
`Broad A bands of striated muscle in Leber's con(cid:173)
`genital amaurosis: A new congenital myopathy?
`R.E. Mrak, B. Lange, and M.C. Brodsky
`
`Guillain-Barre syndrome (GBS) with bilateral
`optic neuritis and central white matter disease
`N. Nadkarni and R.P. Lisak
`
`Clinical/Scientific Note
`
`844
`
`Fulminant rhabdomyolysis in a patient with der(cid:173)
`matomyositis D. V. Caccamo, C. Y. Keene, J. Durham,
`andD. Peuen
`
`846
`
`847
`
`847
`
`848
`
`815
`
`Correspondence
`
`Dropped head syndrome To the Editor: V.J. Lerman.
`To the Editor: T.P. Bleck. To the Editor: S. Khella and
`M. Frederic. Reply from the Authors: G.A. Suarez and
`J .J. Kelly, Jr.
`
`ITP and neuropathy To the Editor: A.J. Giampolo and
`M.J. Giuliani. Reply from the Authors: M.K. Greenberg
`and T. Sonoda
`
`ECT-induced status To the Editor: R.D. Weiner,
`J. Balla, R.A. Radtke, and T.E. Sibert. Reply from the
`Authors: S.I. Lee and N.K. Varma
`
`Secondary syringomyelia To the Editor: M. Sarkarati
`and R .M. Weinstein. Reply from the Authors: A. Feue,
`C. Wallays, M.H. Nicolle, and A. Guillard
`
`Correction
`
`To: Loss of basic fibroblast growth factor in sub(cid:173)
`stantia nigra neurons in Parkinson's disease
`I. Tooyama, T. Kawamata, D. Walker, T. Yamada,
`K. Hanai, H. Kimura, M. !wane, K. Igarashi,
`E.G. McGeer, and P.L. McGeer (February 1993 issue,
`pp. 372-376)
`
`Departments
`
`4A
`llA
`12A
`25A
`849
`
`Information for Authors
`Newsletter
`Calendar
`Professional Notices
`Book Reviews
`
`N eurology uses recycled paper and soy-based ink for its subscription
`and advertising promotion material. As a leading health care publi(cid:173)
`cation, we are committed to contributing to a healthier environment.
`
`PRINTED WITH
`SOY INK
`
`SA NEUROLOGY 43 April 1993
`
`3
`
`

`

`articles
`
`I expedited publication I
`Interferon beta-lb is effective
`in re la psing-reDli tting
`m.ultiple sclerosis.
`I. Clinical results of a multicenter, randomized,
`double-blind, placebo-controlled trial
`
`The IFNB Multiple Sclerosis Study Group*
`
`Article abstract-We report a multicenter, randomized, double-blind, placebo-controlled trial of interferon beta-lb
`(IFNB ) in 372 ambulatory patients with relapsing-remitting multiple sclerosis (MS). Entry criteria included an
`Expanded Disability Status Scale (EDSS) score of 0 to 5.5 and at least two exacerbations in the previous 2 years. One(cid:173)
`third of the patients received placebo, one-third 1.6 million international units (MIU) of IFNB, and one-third 8 MIU of
`IFNB, self-administered by subcutaneous injections every other day. The primary end points were differences in exac(cid:173)
`erbation rates and proportion of patients remaining exacerbation-free. The annual exacerbation rate for patients
`receiving placebo was 1.27; for 1.6 MIU IFNB, 1.17; and for 8 MIU IFNB, 0.84 after 2 years. Exacerbation rates were
`significantly lower in both treatment groups compared with the placebo group (8 MIU versus placebo, p = 0.0001; 1.6
`MIU versus placebo, p = 0.0101; and 8 MIU versus 1.6 MIU, p = 0.0086), suggesting a dosage effect. The reduction in
`exacerbation severity in the 8 MIU group was attributable to a twofold reduction in the frequency of moderate and
`severe attacks. More patients in the 8 MIU group (n = 36) were exacerbation-free at 2 years compared with the placebo
`group (n = 18; p = 0.007 ). EDSS scores changed little from baseline in both the placebo and treatment arms.
`Accordingly, a significant change in disability could not be discerned in this trial. Finally, in serial MRis, MS activity
`was significantly less in the high-dose IFNB group. IFNB treatment was well tolerated: the significant reductions in
`exacerbation rates, severity of exacerbations, and accumulation of MRI abnormalities occurred in the ·absence of seri(cid:173)
`ous side effects. IFNB is the only treatment that has substantially altered the natural history of MS in a properly con(cid:173)
`trolled clinical trial.
`
`NEUROLOGY 1993;43:655-661
`
`Multiple sclerosis (MS) is a common neurologic dis(cid:173)
`ease that is a major cause of disability and econom(cid:173)
`ic loss , especially in young adults . Extensive
`research has failed to identify the cause of MS, but
`present consensus is that damage to the nervous
`system results from immunologic processes. 1•2
`Although many therapies have been used, none has
`significantly reduced exacerbation rates, the accu- .
`mulation of disability, and the increase in MS
`lesion burden as judged by cranial MRI. 3 The
`advent of MRI has facilitated diagnosis and provid(cid:173)
`ed a sensitive index of disease activity and extent
`of disease burden. 4•5 Evaluation of treatment has
`remained difficult because of the well-known ten(cid:173)
`dency for natural remission of symptoms; even
`MRI lesions frequently regress without treatment.
`There have been no previous reports of any treat-
`
`ment capable of preventing both exacerbations and
`new MRI lesions.
`Three rationales existed for trials of interferon
`beta in MS. Jacobs et al, 6 suspecting that MS was
`due to a virus, injected natural interferon beta
`intrathecally and reported a significant reduction
`in exacerbations. Earlier studies used both natural
`and recombinant alpha and beta interferons
`because of their immunomodulatory functions.
`These studies were encouraging but inconclusive. 7•8
`A third rationale for trial of interferon beta in MS
`is provided by two prospective studies9•10 showing
`that intercurrent clinical viral infections trigger
`new attacks and may be the most powerful envi(cid:173)
`ronmental influence on the MS process.11
`A double-blinded, dose-finding pilot study12 in
`subjects with relapsing-remitting MS showed that
`
`I See also pages 641 and 662
`
`'' See page 660 fo r the IFNB Multiple Sclerosis Study Group participants.
`Received Februa ry 6, 1993. Accepted for publication in final form February 12, 1993.
`Address correspondence and reprint requests to Dr. Willia m A Sibley, University Hospital, Tucson, AZ 85724.
`
`April 1993 NEUROLOGY 43 655
`
`r
`
`4
`
`

`

`recombinant interferon beta-lb (IFNB) could be
`administered safely at a dose of 8 million interna(cid:173)
`tional units (MIU) every other day, and suggested
`that treatment decreased the risk of exacerbations.
`This report presents pooled results from two
`identically designed, 2-year, multicenter, double(cid:173)
`blind, placebo-controlled studies, each with three
`parallel treatment groups (placebo, and 1.6 MIU
`and 8 MIU IFNB). One study was conducted in the
`United States and one in Canada. Enrollment
`began in June 1988 and continued to May 1990.
`Data on the first 338 enrollees were subjected to an
`intent-to-treat analysis after 2 years. All subjects
`still in the study were given the option of continu(cid:173)
`ing treatment in a double-blinded fashion, extend(cid:173)
`ing the total treatment period to 57'2 years for some
`patients. The safety and efficacy results for the
`first year of these extension studies are also includ(cid:173)
`ed and discussed as 3-year data.
`
`Methods. Patients. Three hundred seventy-two patients,
`all of whom had had clinically definite or laboratory-sup(cid:173)
`ported definite MS 13 for more than 1 year, were entered
`into the study in 11 different medical centers in the
`United States and Canada. The studies were approved
`by the institutional review boards of the participating
`centers, and all subjects gave informed consent. The dis(cid:173)
`tribution among centers and results of randomization
`into the three· treatment arms of the study are shown in
`table 1. All patients were between the ages of 18 and 50
`years, were ambulatory with Kurtzke Exp a nd e d
`Disability Status Scale (EDSS)14 scores of 5.5 or less, and
`had had at least two acute exacerbations during the pre(cid:173)
`vious 2 years. All had been clinically stable for at least
`30 days before entry and had received no ACTH or pred(cid:173)
`nisone during this period. Prior treatment with azathio(cid:173)
`prine or cyclophosphamide excluded patients from the
`study.
`'
`Exacerbations. An exacerbation was defined as the
`appearance of a new symptom or worsening of an old
`symptom, attributable to MS; accompanied by an appro(cid:173)
`priate new neurologic abnormality; lasting at least 24
`hours in the absence of fever; and preceded by stability
`"or improvement for at least 30 days. Documentation of
`an exacerbation implied that the investigator thought
`there was at least one new MS lesion or enlargement of
`an old one.
`Study medication. IFNB was manufactured by Chiron
`Corporation, Emeryville, CA, and supplied to the investi(cid:173)
`gators as Betaseron by Berlex Laboratories, Richmond,
`CA. Betaseron is a 165-amino acid, 18,500-dalton, type I
`interferon made in Escherichia coli, using recombinant
`technology. Each milligram of IFNB contained 32 MIU of
`IFNB as measured in an assay on human fibroblast
`monolayers, using a vesicular stomatitis virus challenge.
`Identical vials containing lyophilized IFNB (either 1.6
`MIU or 8 MIU)* and placebo were prepared. Each IFNB
`vial contained 15 mg USP human albumin and 15 mg
`dextrose in addition to .the IFNB content, and each pla-
`
`* Subsequent to the conduct of this study, the standard for potency of the
`international units of IFN beta have been changed. The previous refer(cid:173)
`ence standard (native IFNB, National Institutes of Health) indicated the
`dosages to be 9 million IU (9 MIU) a nd 45 million IU (45 MIU). The new
`international reference standard (r ecombinant IFNB , World Healt h
`Organization ) translates the dosages to 1.6 MIU and 8 MIU, respectively.
`This publication uses the latter standard.
`
`656 NEUROLOGY 43 April 1993
`
`Table 1. Enrollment of patients by study center
`
`Site
`
`TempleU
`Thomas J efferson
`UCSF
`U Alabama
`U Arizona
`U Chicago
`U Maryland
`H6pital de Notre Dame
`Montreal Neurologic Inst.
`U British Columbia
`U Hospital , London, Ont.
`
`Placebo
`
`IFNB
`1.6 MIU 8MIU Total
`
`10
`10
`IO
`4
`12
`16
`10
`12
`8
`17
`14
`
`10
`10
`10
`4
`12
`16
`10
`12
`8
`18
`15
`
`10
`10
`10
`4
`11
`16
`10
`12
`8
`17
`16
`
`30
`30
`30
`12
`35
`48
`30
`36
`24
`52
`45
`
`Totals
`
`123
`
`125
`
`124
`
`372
`
`cebo vial contained only a similar quantity of albumin
`and dextrose.
`Withdrawal criteria. These were ( 1) steady progres(cid:173)
`sion of disability for 6 months, (2) treatment with more
`than three courses of ACTH or prednisone (each 28 days
`or less) during a 1-year period, (3) failure to take sched(cid:173)
`uled doses of trial medication for more than 2.consecu(cid:173)
`tive weeks, and (4) moderate or severe drug toxicity per(cid:173)
`sisting on rechallenge.
`Study design. After randomization, each patient was
`instructed in self-administration of study medication .
`The first three .injections were given at the study center
`under observation ; th ereafter, patients injected them(cid:173)
`selves subcutaneously every other day.
`All personnel at each study site were blinded to treat(cid:173)
`ment categories. Two physicians at each site were desig(cid:173)
`nated: one neurologist who was not aware of drug side
`effects to do the periodic examinations, and another
`"treating" neurologist who knew about side effects and
`injection site reactions, reviewed laboratory findings for
`toxicity, and was responsibl e for overall care. When sig(cid:173)
`nificant toxicity occurred, study medication was stopped,
`and resumed at one-half dose when manifestations sub(cid:173)
`sided. If toxicity recurred, the patient was dropped from
`the study; if not, the one-half dosage was continued for
`the balance of the trial.
`After the first few months of more frequent visits,
`patients were evaluated every 12 weeks, or more often if
`symptoms occurred suggesting the possibility of MS
`exacerbation. Each evaluation included a standard neu(cid:173)
`rologic examination, and a Scripps Neurologic Rating
`Scale (NRS) 15 score and a Kurtzke EDSS score were
`determined.
`Each patient had a baseline cranial MRI, and thi s was
`repeated yearly. A cohort of 52 patients at the University
`of British Columbia also had head MRis repeated at 6-
`week intervals for 2 years. Techniques for measurement
`of the total area of MS involvement by MRI as well as
`details of the care taken to ensure consistent scans at all
`study sites are presented in a separate publication. 16 All
`MRis were interpreted by the Radiology Department of
`the University of British Columbia in a blinded fashion.
`Efficacy criteria. The primary end points of the stud(cid:173)
`ies were annual exacerbation rate and proportion of
`exacerbation-free patients. Secondary end points includ(cid:173)
`ed time to first exacerbation (days); exacerbation dura(cid:173)
`tion and severity (as defined by quantitative change in
`NRS score: 0 to 7 = mild ; 8 to 14 = moderate; and >15 =
`severe), change in EDSS and NRS scores from those at
`
`r
`
`5
`
`

`

`I
`
`·[ Table 2. Demographic and baseline disease
`characteristics of all enrollees (372 subjects)
`
`Table 3. Effect of IFNB on exacerba tions:
`Summary of 2- and 3-year data
`
`Variable
`
`Placebo
`(n = 123)
`
`IFNB
`1.6 MIU
`8MIU
`(n = 125)
`(n = 124)
`
`Two-year data
`
`Exacerbation rate
`Exacerbation-free subjects
`Median time to first
`exacerbation (days)
`
`Three-year data
`
`Exacerbation rate
`Exacerbation-free subjects
`Median time to first
`exacerbation (days )
`
`Placebo
`(n = 112)
`
`IFNB
`1.6 MIU
`SMIU
`(n = lll) (n = 115)
`
`1.27
`18
`153
`
`1.17
`23
`180
`
`0.84*
`36
`295t
`
`p value*
`
`0.0001
`0.007
`0.015
`
`Placebo
`(n = 123)
`
`IFNB
`1.6 MIU
`8MIU
`(n = 125)
`(n = 124)
`
`1.21
`17
`147
`
`1.05
`23
`199
`
`0.84
`27
`264
`
`p value*
`
`0.0004
`0.097 NS
`0.028
`
`MIU Million international units.
`• p < 0.01, 8 MIU vs 1.6 MIU.
`p < 0.05, 8 MIU V S 1.6 MIU.
`p value for 8 MIU vs placebo.
`
`second exacerbations. Data on patients were censored at
`the time of withdrawal. The log-rank statistic was used
`to test for comparability of the survival curves for each
`treatment arm.
`
`Results. Demographic and baseline characteristics
`in the three treatment arms after randomization.
`The three groups were comparable (table 2). In
`each there was an approximate 2:1 female-to-male
`preponderance and nearly the same mean age, dis(cid:173)
`ease duration, and baseline EDSS. Moreover, all
`three groups had had roughly the same number of
`exacerbations in the 2 years before the study.
`Effect of IFNB on primary outcome measures.
`The primary and secondary efficacy results at 2
`years (table 3) revealed that the exacerbation rate
`was 1.27 per year in the placebo group, 1.17 for the
`1.6 MIU group, and 0.84 for the 8 MIU group
`(placebo versus 8 MIU, p = 0.0001; 1.6 MIU versus
`8 MIU, p = 0.0086, and 1.6 MIU versus placebo, p =
`0.01) (figure 1). The number of patients exacerba(cid:173)
`tion-free was 18, 23, and 36, respectively, at 2
`years (placebo versus 8 MIU, p = 0.007; 1.6 MIU
`versus 8 MIU, p = 0.076, NS) (figure 2).
`Secondary outcome measures. There was a sig(cid:173)
`nificant prolongation of time to first and second
`exacerbations for the 8 MIU group compared with
`placebo after 2 years of treatment (table 3). This
`beneficial effect continued after 3 years of treat(cid:173)
`ment, at which time the exacerbation rates
`remained significantly different for placebo versus
`8 MIU (p = O.Q004), and the time to first and fifth
`exacerbations were also significantly prolonged.
`The number of patients exacerbation-free, however,
`which had been significantly different after 2 years
`of treatment with 8 MIU, was no longer significant(cid:173)
`ly different after 3 years of treatment (placebo ver(cid:173)
`sus 8 MIU, p = 0.097, NS).
`Exacerbations were rated as either mild, moder(cid:173)
`ate, or severe according to changes in the NRS
`score. Table 4 shows that the combined annualized
`rates of moderate and' severe exacerbations in the 8
`
`April 1993 NEUROLOGY 43 657
`
`88
`35
`
`116
`7
`
`Sex
`Female
`Male
`Race
`White
`Other
`Age (yr)
`Mean
`SEM
`Age at diagnosis (yr)
`32.1
`Mean
`0.6
`SEM
`MS duration from diagnosis (yr)
`3.9
`Mean
`0.3
`SEM
`Baseline EDSS
`Mean
`SEM
`Baseline NRS
`81.1
`Mean
`1.0
`SEM
`Exacerbations in prior 2 years
`Mean
`3.6
`SEM
`0.1
`Last exacerbation before entry (days)
`Mean
`134.9
`SEM
`10.3
`
`36.0
`0.6
`
`2.8
`0.1
`
`85
`40
`
`116
`9
`
`35.3
`0.7
`
`30.6
`0.7
`
`4.7
`0.4
`
`2.9
`0.1
`
`80.8
`1.1
`
`3.3
`0.1
`
`86
`38
`
`116
`8
`
`35.2
`0.6
`
`30.5
`0.6
`
`4.7
`0.4
`
`3.0
`0.1
`
`80.6
`1.0
`
`3.4
`0.2
`
`140.8
`10.2
`
`157.9
`10.8
`
`MIU Million international units.
`There were no significant differences between any of these groups.
`
`baseline; quantitative disease burden as measured by
`annual MRI; and disease activity as measured by MRI in
`· a frequent-scanning substudy.
`Statistical methods. As specified in the study proto(cid:173)
`cols, the data from the two studies were combined and
`analyses were based on an intent-to-treat data set. The
`analyses for the extension studies covered the full 3
`years of available data. SAS software was used for all
`analyses.
`For continuous variables, treatment-group differences
`were analyzed using an analysis of variance (ANOV A)
`based on ranked data. This model accounted for the fol(cid:173)
`lowing effects: treatment group, study site, and treat(cid:173)
`ment group by study site. Categorical variables, such as
`sex and race, were tested for general association .by use
`of the Cochran-Mantel-Haenszel (CMH) chi-square test
`stratified for site. Similarly, categorical variables with an
`ordinal scale, such as exacerbation count per patient,
`were tested for treatment group differences using the
`CMH ANOV A statistic. This statistic corresponds to a
`site-adjusted ANOVA. A two-tailed Fisher's exact test
`was used for categorical data that were sparse. In the
`display of group exacerbation rates, the 95% confidence
`intervals were calculated using a Poisson distribution
`based on the number of observed exacerbations in each
`treatment group.
`Survival curves were calculated with life-table meth(cid:173)
`ods for the length of time before the onset of the first and
`
`6
`
`

`

`MIU group at 2 years equal~d 0.23, one-half the
`0.45 rate for moderate and severe exacerbations in
`the placebo group (8 MIU' \tersus placebo, p =
`0.002). Sometimes patients were unable to visit the
`center promptly after exacerbations and had
`already improved to some extent when first exam(cid:173)
`ined; in such instances, the NRS score could not be
`calculated at the height of the attack (the
`"unknowns" in table 4).
`One effect of the decreased number of severe
`exacerbations in the high-dose group was less need
`
`p=0.0001
`
`p=0.0101
`
`p=0.0086
`
`II
`
`61 .33
`1.17
`1.03
`
`I
`
`I
`
`81 .43
`1.27
`1.12
`
`I
`
`I
`
`80.97
`0.84
`.· 0.72
`
`2.0 -
`
`Q)
`
`iii
`a:
`c:
`0
`~ 1.5
`....
`.D
`Q)
`()
`cu
`x w 1.0 -
`co ::i
`c:
`c: <
`
`0.5
`
`0
`
`Placebo
`
`1.6 million IU
`
`Smillion IU
`
`Figure 1. Exacerbation rates for the three treatment
`groups (boldface) in the first 2 years of the study, and
`95% confidence intervals. The significance of the
`reduction in the 8 MIU group was essentially unchanged
`in the 3-year analysis. Overall, p = 0.0001 .
`
`100
`
`90
`
`80
`
`for hospitalization (table 5)
`MRI lesion area. At the end of the first year,
`mean lesion area in relation to baseline, by MRI,
`increased by 12.2% in the placebo group and by
`
`Table 4. Severity of exacerbations by treatment
`group (2-year data)
`
`Placebo
`(n = 112)
`
`IFNB
`1.6 MIU
`SMIU
`(n = 111) (n = 115)
`
`Patients with exacerbations
`Number of exacerbations
`Total patient-years on study
`Annual rate
`Mild
`Moderate and severe
`Unknown
`Totals
`
`94
`266
`209.2
`
`0.54
`0.45
`0.28
`1.27
`
`88
`242
`207 .0
`
`0.62
`0.32
`0.22
`1.17
`
`79
`173
`207.0
`
`0.45
`0.23
`0.15
`0.84
`
`Placebo vs 8 MIU for moderate/severe, p = 0.002 .
`Overall for moderate/severe, p = 0.007.
`
`Table 5. Hospitalizations for MS during 3 years of
`study
`
`IFNB
`Placebo l.6 MIU B MIU p va lue
`
`Num ber of hospitalizations
`
`65
`
`Number of patients
`hospitalized
`
`33
`
`53
`
`25
`
`37
`
`21
`
`Number of hospital stays
`
`47 1
`
`411
`
`344
`
`0.046
`!placebo vs B MIUJ
`
`0.129 (overall I
`0.05
`(placebo vs B MIUJ
`
`0.07 (overall l
`0.023
`!placebo vs 8 MIU J
`
`70
`
`60
`
`g
`:0 cu
`.D e 50
`a..
`~ 0
`
`40
`
`30
`
`20
`
`10
`
`0
`
`-·'"1- ... ,,
`'· -----·"··-------·
`'··: ..... __ __ "··
`.... _______ ,
`"''"'-----· ..........
`----·----·-·-··--"--., __
`
`'------·---·-
`
`0
`
`60
`
`120
`
`180
`
`240
`
`420
`360
`300
`Time (days)
`
`480
`
`540
`
`600
`
`660
`
`720
`
`Group
`
`- - Placebo
`
`--- 1.6 million IU
`
`------· 8 million IU
`
`658 NEUROLOGY 43 April 1993
`
`Figure 2. Kaplan-Meier
`analysis showing the
`probability of remaining
`exacerbation-free in the
`first 2 years of the study.
`
`(
`
`I
`I
`I
`
`I
`
`7
`
`

`

`(
`I
`
`4.1 % in the intermediate-dose IFNB group, and
`decreased by 1.1 % in the high-dose group. At the
`end of the second year, the numbers were 20%
`increase (placebo group), 10.5% increase (1.6 MIU
`group), and 0.1 % decrease in the high-dose group.
`At the end of 3 years the figures were ;·17 .1 %
`increase (placebo), 1.1% increase (1.6 MIU), and a
`6.2% decrease in the high-dose group of patients.
`Additional MRI data are contained in a companion
`publication in this issue. 16
`·
`EDSS and NRS. The NRS score, an attempt to
`express the neurologic examination as a single
`number, showed no material change in any of the
`t reatment groups in any of the 3 study years.
`Likewise, there was no significant change in the
`mean EDSS score by treatment arm, although the
`numbers were all slightly higher in all three groups
`at 3 years compared with baseline. There was a
`trend suggesting lessened disability at the 3-year
`end point in the 8 MIU group (p = 0.043) (table 6).
`"End point" in table 6 denotes the EDSS score at
`the end of 3 years (or the EDSS at time of dropout);
`"confirmed end point" indicates two consecutive
`EDSS scores, separated by 90 days, that were iden(cid:173)
`tical, with both showing a 1.0 point increase over
`the baseline score. To calculate confirmed end
`point, use of early fourth-year examination data
`was necessary in most patients.
`Withdrawals, adverse events. and side effects.
`Sixty-five patients discontinued treatment during
`the first 2 years (23 in the placebo, 18 in the 1.6
`MIU, and 24 in the 8 MIU groups). Although 122
`patients did not finish the third year, that figure
`included 22 who chose not to continue when the
`study was extended. The overall dropout rate did
`not exceed that anticipated at the trial's inception.
`Moreover, the distribution of dropouts remained
`similar in the three treatment arms although the
`reasons were different. More placebo patients with(cid:173)
`drew for lack of efficacy; there were 10 withdrawals
`in this group at 2 years for excessive use of steroids
`compared with only one withdrawal in each of the
`IFNB treatment arms for this reason. In contrast,
`more IFNB-treated patients discontinued because
`of toxicity. Among the 16 patients withdrawing for
`treatment side effects at 2 years, one was in the
`placebo group, five in the 1.6 MIU arm, and 10 in
`the high-dose group. Only one additional adverse(cid:173)
`event withdrawal, a suicide, occurred in the third
`treatment year. The reasons for withdrawal due to
`treatment effects (patient numbers in parentheses)
`were abnormal liver enzymes (3), injection site pain
`(3), fatigue (3), and cardiac arrhythmia, allergic
`reaction, nausea, headache, "flu syndrome," confu(cid:173)
`sion, and "felt sick" ( 1 each).
`Other causes of withdrawal included unblinding
`events (9), excessive use of steroids ( 14), major non(cid:173)
`compliance (2), exacerbation at baseline (1), and
`emotional instability (1). In addition, there was a
`wide variety of personal reasons, including dislike
`of injections, moving, and a desire to become preg(cid:173)
`nant.
`
`Table 6. Effect of IFNB treatment on disability at
`3-year end point
`
`End point•
`Stable Worsenedt
`
`Confirmed end point•
`Stable
`Worsenedt
`
`Placebo
`IFNB
`l.6MIU
`8 MIU
`Placebo vs 8 MIU
`
`74 (61%)
`
`48(39%)
`
`88 (72%)
`
`34 (28%)
`
`81 (65%)
`44 (35%)
`33 (27%)
`89 (73%)
`p = 0.043
`
`90 (72%)
`35 (28%)
`25 (20%)
`97 (80%)
`p = 0.161
`
`• Defin ed in text.
`t Worsened by at least one EDSS point.
`
`The most common laboratory abnormality asso(cid:173)
`ciated with "IFNB treatment was a mild intermit(cid:173)
`tent lymphopenia unassociated with significant
`changes in the total white blood cell count.
`Lymphopenia occurred at some time in 65% of
`placebo patients, 76% in the 1.6 MIU group, and
`80% of the 8 MIU patients. Mild neutropenia
`occurred in 4% of the placebo, 6% of the 1.6 MIU,
`and 17 % of the 8 MIU groups. There were no
`severe neutropenias.
`Although mild neutropenia, anemia, and throm(cid:173)
`bocytopenia were present and statistically related
`to treatment group (being more common in the 8
`MIU group), they were clinically insignificant and
`also sporadic. Mild or moderate SGPT changes
`occurred in five placebo patients, seven 1.6 MIU
`patients, and 14 high-dose patients at some time.
`Only four patients had elevations of SGOT.
`Overall, incidence of notable side effects was
`low. Injection site reactions were seen in all treat(cid:173)
`ment groups but more commonly in the 8 MIU
`group.