art ides
`
`I expedited publication 1
`Copolymer 1 reduces
`relapse rate and imwoves
`disability in relapsing-remitting
`multiple sclerosis:
`Results of a phase I11 multicenter, double-blind,
`placebo-controlled trial
`
`K.P. Johnson, MD; B.R. Brooks, MD; J.A. Cohen, MD; C.C. Ford, MD; J. Goldstein, MD; R.P. Lisak, MD;
`L.W. Myers, MD; H.S. Panitch, MD; J.W. Rose, MD; R.B. Schiffer, MD; T. Vollmer, MD; L.P. Weiner, MD;
`J.S. Wolinsky, MD; and the Copolymer 1 Multiple Sclerosis Study Group"
`
`Article a b s t r a c t w e studied copolymer 1 (Copaxone) in a multicenter (11-university) phase I11 trial of patients with re-
`lapsing-remitting multiple sclerosis (MS). Two hundred fifty-one patients were randomized to receive copolymer 1 (n =
`125) or placebo (n = 126) at a dosage of 20 mg by daily subcutaneous injection for 2 years. The primary end point was a
`difference in the MS relapse rate. The final 2-year relapse rate was 1.19 * 0.13 for patients receiving copolymer 1 and
`1.68 * 0.13 for those receiving placebo, a 29% reduction in favor of copolymer 1 (p = 0.007) (annualized rates = 0.59 for
`copolymer 1 and 0.84 for placebo). Trends in the proportion of relapse-free patients and median time to first relapse fa-
`vored copolymer 1. Disability was measured by the Expanded Disability Status Scale (EDSS), using a two-neurologist (ex-
`amining and treating) protocol. When the proportion of patients who improved, were unchanged, or worsened by 21 EDSS
`step from baseline to conclusion (2 years) was evaluated, significantly more patients receiving copolymer 1 were found to
`have improved and more receiving placebo worsened (p = 0.037). Patient withdrawals were 19 (15.2%) from the copolymer
`1 group and 17 (13.5%) from the placebo group at approximately the same intervals. The treatment was well tolerated.
`The most common adverse experience was an injection-site reaction. Rarely, a transient self-limited systemic reaction fol-
`lowed the injection in 15.2% of those receiving copolymer 1 and 3.2% of those receiving placebo. This reaction was charac-
`terized by flushing or chest tightness with palpitations, anxiety, or dyspnea and commonly lasted for 30 seconds to 30
`minutes. This rigorous study confirmed the findings of a previous pilot trial and demonstrated that copolymer 1 treat-
`ment can significantly and beneficially alter the course of relapsing-remitting MS in a well-tolerated fashion.
`NEUROLOGY 1995;45: 1268-1276
`
`Progress in identifying effective therapies for multi-
`copolymer 1 (Copaxone), given subcutaneously (s.c.)
`ple sclerosis (MS) has accelerated during this decade
`at a dosage of 20 mg per day in a rigorously con-
`as pathogenic factors active in the disease have been
`trolled 2-year trial, significantly reduced the relapse
`identified. We now report that t r e a t m e n t with
`rate in patients with relapsing-remitting MS. Neuro-
`I
`I
`Seealsopage 1245
`
`*See pages 1275 and 1276 for the Copolymer 1 Multiple Sclerosis Study Group participants.
`From the Department of Neurology (Drs. Johnson and Panitch), University of Maryland, Baltimore, MD; the Department of Neurology (Dr. Brooks), Uni-
`versity of Wisconsin, Madison, WI; the Department of Neurology (Dr. Cohen), University of Pennsylvania, Philadelphia, P A the Department of Neurol-
`ogy (Dr. Ford), University of New Mexico, Albuquerque, NM; the Department of Neurology (Drs. Goldstein and Vollmer), Yale University, New Haven,
`CT; the Department of Neurology (Dr. Lisak), Wayne State University, Detroit, MI; the Department of Neurology (Dr. Myers), University of California,
`Los Angeles, CA; the Department of Neurology (Dr. Rose), University of Utah and the Veterans Administration Medical Center, Salt Lake City, UT; the
`Department of Neurology (Dr. Schiffer), University of Rochester, Rochester, NY; the Department of Neurology (Dr. Weiner), University of Southern Cali-
`fornia, Los Angeles, CA; and the Department of Neurology (Dr. Wolinsky), University of Texas, Houston, TX.
`Supported by the Federal Food and Drug Administration Orphan Drug Program no. FD-R000559-01, the National Multiple Sclerosis Society no. RG
`2202-A-6, and Teva Pharmaceutical Industries, Ltd., Petah Tiqva, Israel.
`Presented at the annual meeting of the American Neurological Association, San Francisco, October 1994.
`Received April 27, 1995. Accepted in final form May 1, 1995
`Address correspondence and reprint requests to Dr. Kenneth P. Johnson, Department of Neurology, N4W46, University of Maryland Hospital, 22 South
`Greene Street, Baltimore, MD 21201.
`
`1268 NEUROLOGY 45 July 1995
`
`Page 1 of 9
`
`YEDA EXHIBIT NO. 2004
`MYLAN PHARM. v YEDA
`IPR2017-00195
`
`

`

`logic impairment, as measured by the Expanded
`Disability Status Scale (EDSS),I was also favorably
`affected, and patients tolerated treatment well, with
`a low frequency of side effects. Thus, copolymer 1
`joins interferon beta-lb (IFNB-lb) (licensed in 1993)
`as a treatment shown to positively alter the natural
`course of relapsing-remitting MS2
`Copolymer 1 is the acetate salt of a mixture of
`synthetic polypeptides composed of four amino
`acids, L-alanine, L-glutamic acid, L-lysine, and L-ty-
`rosine, in a molar ratio of 4.2, 1.4, 3.4, and 1.0, re-
`spectively, and with an average molecular weight
`of 4,700 to 13,000 daltons. First synthesized in
`1967 by M. Sela, R. Arnon, D. Teitelbaum, and
`their colleagues at the Weizmann Institute of Sci-
`ence in Israel, copolymer 1 suppresses or modifies
`experimental allergic encephalomyelitis
`in
`several species of mammals including nonhuman
` primate^.^ Other studies5 suggest that copolymer 1
`acts through cross-reactivity with myelin basic pro-
`tein (MBP) and inhibition of the cell-mediated im-
`mune response to this antigen.
`Extensive preclinical findings encouraged
`Abramsky et a16 to treat a small number of patients
`who had advanced MS or acute disseminated en-
`cephalomyelitis with copolymer 1. They used a low
`dose and observed no toxicity. Bornstein et a17 then
`treated four MS patients in the relapsing-remitting
`and 12 in the chronic-progressive stages of disease
`with copolymer 1 and noted fewer relapses or neu-
`rologic improvement in five. They used various
`doses and routes of administration for up to 6
`months. This open trial was later extended and the
`dose increased from 5 mg i.m. to 20 mg S.C. daily for
`up to 3 years without significant side effects or lab-
`oratory abnormalities.
`These early human studies indicated t h a t
`copolymer 1 could be given safely and prompted a
`2-year, placebo-controlled, double-blind pilot trial
`to evaluate its effects on the MS relapse rate, dis-
`ability, and patient tolerance.8 Forty-eight patients
`with relapsing-remitting MS, a high mean annual
`relapse rate of 1.9, and a mean disability status
`scale (EDSS) score of 3.0 were entered. Twenty-five
`received 20 mg of copolymer 1 S.C. daily and 23 re-
`ceived S.C. placebo. During 2 years, there were 62
`relapses in the placebo group but only 16 in the
`copolymer 1 group, a highly significant difference.
`Fifty-six percent of the copolymer 1 group and 26%
`of those receiving placebo remained relapse-free.
`The effect was most pronounced in patients with
`the lowest EDSS ratings at entry, and there was a
`trend toward benefit of copolymer 1 over placebo in
`terms of progression of disability, especially in the
`patients with the lower EDSS scores at entry. Pa-
`tient tolerance was very good, and there were no
`laboratory abnormalities.8
`Copolymer 1 was then studied in patients with
`chronic-progressive MS at two centers, the Albert
`Einstein College of Medicine, Bronx, NY, and the
`Baylor College of Medicine, Houston, TX.9 Patients
`with EDSS ratings from 2.0 to 6.5, inclusive, were
`
`Table 1. Participating universities and the number
`of patients randomized to each treatment group
`
`Center
`
`Copolymer 1 Placebo
`
`University of California,
`Los Angeles
`University of Maryland*
`University of New Mexico
`University of Pennsylvania
`University of Rochester
`University of Southern
`California
`University of Texas, Houston
`University of Utah
`Wayne State University
`University of Wisconsin
`Yale University
`* National coordinating center.
`
`16
`
`14
`13
`14
`15
`6
`
`9
`12
`12
`6
`8
`
`14
`
`14
`14
`13
`13
`8
`
`11
`12
`12
`7
`8
`
`observed for at least 12 months before randomiza-
`tion to document progression of their disease. One
`hundred six patients (mean age 42 years, mean
`EDSS score 5.6) were treated in a double-blind
`trial. They received either placebo or 15 mg of
`copolymer 1 twice daily by S.C. self-injection, and
`tolerated the therapy well. The combined results
`showed a trend toward benefit with copolymer 1
`treatment, which was, however, not statistically
`~ignificant.~
`To further evaluate copolymer 1 treatment of pa-
`tients with relapsing-remitting MS, we conducted a
`large, placebo-controlled, multicenter trial and have
`observed patients in a blinded fashion for 2 years.
`
`Methods. The objectives of the current study were to
`compare the patient tolerance and therapeutic impact of
`daily S.C. injections of 20 mg of copolymer 1 or placebo
`over 24 months, using the number of MS relapses as the
`primary variable. The study was designed and the pa-
`tients recruited to confirm the conclusions of the previ-
`ously published pilot triaL8
`Participants. Eleven universities with active MS cen-
`ters and experience in conducting clinical neurologic re-
`search participated in the trial (table 1). The University
`of Maryland served as the administrative and clinical co-
`ordinating center. After an intensive training session for
`neurologists and study coordinators, the trial began in
`October 1991.
`Study design. The primary end point, determined
`prospectively in this phase I11 study, was a comparison of
`the mean number of relapses experienced by copolymer
`1- or placebo-treated relapsing-remitting MS patients
`during 2 years of treatment. A relapse was defined as the
`appearance or reappearance of one or more neurologic
`abnormalities persisting for at least 48 hours and imme-
`diately preceded by a relatively stable or improving neu-
`rologic state of at least 30 days. A relapse was confirmed
`only when the patient’s symptoms were accompanied by
`objective changes on the neurologic examination consis-
`tent with an increase of at least a half a step on the
`EDSS, two points on one of the seven functional
`systems,l or one point on two or more of the functional
`
`July 1995 NEUROLOGY 45 1269
`
`Page 2 of 9
`
`YEDA EXHIBIT NO. 2004
`MYLAN PHARM. v YEDA
`IPR2017-00195
`
`

`

`systems. Events associated with fever were excluded. A
`change in bowellbladder or cognitive function could not
`be solely responsible for the changes in either the EDSS
`or the functional system scores. Several secondary end
`points were also predetermined: proportion of relapse-
`free patients, time to first relapse after initiation of ther-
`apy, proportion of patients with sustained disease pro-
`gression (defined as an increase of at least one full step
`on the EDSS that persisted for a t least 3 months), and
`mean change in EDSS and ambulation index between
`the copolymer 1 and placebo groups from baseline to con-
`clusion. All patients had periodic, standardized neu-
`ropsychological tests, and a subset of patients underwent
`serial gadolinium-enhanced MRIs; results will be re-
`ported in separate publications.
`Conduct of the study. Patients were screened to deter-
`mine eligibility and then randomized within 21 days. A
`centralized randomization scheme was used. All patients
`met the criteria of clinically definite MS or laboratory-
`supported definite MS.’O Male and female patients be-
`tween the ages of 18 and 45 years were eligible. They
`were all ambulatory with an EDSS score of 0 through
`5.0, a history of at least two clearly identified and docu-
`mented relapses in the 2 years prior to entry, onset of the
`first relapse at least 1 year before randomization, and a
`period of neurologic stability and freedom from cortico-
`steroid therapy of at least 30 days prior to entry. Pa-
`tients were excluded if they had ever received copolymer
`1 or previous immunosuppressive therapy with cytotoxic
`chemotherapy (azathioprine, cyclophosphamide, or cy-
`closporine) or lymphoid irradiation. Other exclusion cri-
`teria included pregnancy or lactation, insulin-dependent
`diabetes mellitus, positive HIV or HTLV-I serology, evi-
`dence of Lyme disease, or required use of aspirin o r
`chronic nonsteroidal anti-inflammatory drugs during the
`course of the trial. All women were required to use an ad-
`equate method of contraception.
`The study medication was supplied by Teva Pharma-
`ceutical Industries, Ltd, Petah Tiqva, Israel, under a
`manufacturing protocol approved by the US Food and
`Drug Administration. It was distributed to each of the
`11 cooperating university centers by an independent
`data management and coordination center, National
`Medical Research Corporation, Hartford, CT. Study
`medication was supplied i n single-dose vials of
`lyophilized material along with ampules of sterile water
`diluent. Patients were given a 1-month supply each
`month and were instructed in reconstitution and S.C.
`self-administration of the study medication. At each
`monthly visit, patients received medication and reported
`adverse events and use of concomitant medications.
`Every 3 months, the patients underwent a complete
`evaluation that employed a two-neurologist protocol.
`Each patient was assigned a single examining neurolo-
`gist who evaluated only the objective neurologic condi-
`tion without discussing symptoms or side effects. A sec-
`ond treating neurologist evaluated symptoms and ad-
`verse events and was responsible for determining the
`need for steroid therapy at the time of a confirmed re-
`lapse. A nurse coordinator at each center distributed
`medication, noted concomitant treatments, and obtained
`blood and urine specimens for laboratory analysis. The
`nurse coordinator and both neurologists were blinded to
`study medication assignment throughout the trial. Pa-
`tients were allowed to use the conventional medications
`they were receiving a t the time of randomization for
`spasticity, bladder control, fatigue, and other MS symp-
`toms. An approved protocol for steroid therapy was em-
`
`1270 NEUROLOGY 45 July 1996
`
`ployed by the treating neurologist at the time of con-
`firmed relapse. Use of immunosuppressive, cytotoxic, or
`experimental drugs or aspirin and chronic nonsteroidal
`anti-inflammatory drugs were proscribed.
`At the time of suspected relapse, patients were in-
`structed to call their center immediately to discuss symp-
`toms with the nurse coordinator or treating neurologist
`and to arrange for an examination at the center within 7
`days. In rare instances, weather conditions and other
`emergencies prohibited evaluation at the site within that
`time. Patients were followed as often as medically indi-
`cated after each confirmed relapse.
`All patients had a chest x-ray and ECG at the screen-
`ing visit and another ECG at the conclusion of the study.
`Urinalysis, hematologic studies, a serum chemistry
`panel, and anti-copolymer 1 antibodies were evaluated at
`3-month intervals; all blood testing was done at a cen-
`tralized laboratory and reported to the treating neurolo-
`gist and to the data management and coordination ten-
`ter. An independent safety monitoring committee, com-
`posed of two clinical neurologists, a clinical pharmacolo-
`gist, a statistician, and a representative of the National
`Multiple Sclerosis Society, met quarterly either in person
`or by conference call to review all safety information. At
`no time were representatives of the sponsor or the 11
`study centers present when safety data or issues were
`discussed. The safety committee remained blinded
`throughout the course of the trial.
`The protocol was approved by the institutional review
`boards of the participating clinical centers, and all pa-
`tients gave written informed consent.
`Statistical analysis. The final data set was evalu-
`ated using several cohort definitions. The intention-to-
`treat analysis of all randomized patients was consid-
`ered primary. Other evaluated cohorts excluded pa-
`tients who did not complete 6 months of treatment, pa-
`tients who failed t o complete 2 years (730 days) of
`treatment, and patients who missed over 5% of consec-
`utive study medication doses or 10% of total doses dur-
`ing the study. There was strong internal consistency of
`statistically significant findings and trends among the
`various evaluated cohorts. Therefore, only the results
`of the most rigorous intention-to-treat analysis are pre-
`sented here.
`The proportions of withdrawals were compared using
`the Cochran-Mantel-Haenszel test. Time to withdrawal
`was analyzed using the log rank test. For demographic
`and medical history characteristics, two-sample t tests
`were used for continuous variables and exact probability
`tests for discrete variables.
`Mean relapse rate was analyzed using ANCOVA, with
`tests for study-drug-by-center interaction and including
`a priori-defined covariates: sex, duration of disease
`(years), prior 2-year relapse rate, and baseline Kurtzke
`EDSS. Proportions of relapse-free patients were tested
`using logistic regression incorporating the same covari-
`ate effects. Time to first relapse was evaluated using
`Weibull regression. The proportion of progression-free
`patients was analyzed using logistic regression.
`Changes from baseline for the Kurtzke EDSS and
`the ambulation index were assessed using repeated-
`measures ANCOVA. Analyses of the change from base-
`line to 24 months were also conducted. Categorical re-
`peated-measures and 24-month end-point analyses
`were performed on Kurtzke EDSS score changes from
`baseline, classified as “improved” (reduction of a t least
`one step) “worsened,” (increase of at least one step), or
`“no change.”
`
`Page 3 of 9
`
`YEDA EXHIBIT NO. 2004
`MYLAN PHARM. v YEDA
`IPR2017-00195
`
`

`

`Results. Baseline characteristics of subjects. Be-
`tween October 1991 and May 1992, 284 patients
`were screened and 251 randomized to the two treat-
`ment groups. The demographics of the randomized
`cohort are shown in table 2. The two groups were
`well matched for age, sex, race, duration of disease,
`mean relapse rate in the prior 2 years, EDSS, and
`ambulation index. As expected, the majority of ran-
`domized patients were women (73%) and white
`(94%). Among the patients randomized to receive
`copolymer 1, 51 were in the 0 to 2, 57 in the 2 to 4,
`and 17 in the >4 EDSS range. Of those randomized
`to receive placebo, 68 were in the 0 to 2, 46 in the 2
`to 4, and 12 in the >4 EDSS range.
`Patient exposure and withdrawals. The total pa-
`tient exposure and duration of treatment is shown
`in table 3. The total patient exposure to copolymer
`1 was 227 years and to placebo 232 years. Nineteen
`patients (15%) withdrew from the copolymer 1-
`treated group and 17 (13.5%) from the placebo
`
`Table 2. Demographics and MS characteristics at
`baseline (number screened = 284)
`
`Copolymer 1
`(n = 125)
`
`Placebo
`(n = 126)
`
`Age (yr; mean i SD)
`Sex
`Women
`Men
`Race
`White
`Other
`Prior 2-year relapse rate
`(mean * SD)
`EDSS (mean * SD)
`Ambulation index
`(mean i SD)
`Duration of MS
`(yr; mean * SD)
`EDSS Expanded Disability Status Scale.
`
`34.6 i 6.0
`
`88 (70.4%)
`37 (29.6%)
`
`118 (94.4%)
`7 (5.6%)
`2.9 * 1.3
`2.8 * 1.2
`1.2 * 1.0
`7.3 * 4.9
`
`34.3 i 6.5
`
`96 (76.2%)
`30 (23.8%)
`
`118 (93.6%)
`8 (6.3%)
`2.9 i 1.1
`2.4 * 1.3
`1.1 * 0.9
`6.6 * 5.1
`
`group. The proportion of patients who withdrew
`and the time to withdrawal as shown in table 3
`were statistically similar over the duration of the
`study. Three patients in the copolymer 1 group
`withdrew when they became pregnant, and one
`stopped medication because of disease progression.
`Two patients in the placebo group failed to comply
`with the protocol. Two copolymer 1 patients with-
`drew for serious adverse events: one, after 50 days
`on treatment, developed immediate flushing, chest
`tightness, dyspnea, nausea, and vomiting (see
`below), which lasted for more than 90 minutes
`after the injection, and one, after 131 days on treat-
`ment, developed generalized lymph node enlarge-
`ment. Lymph node biopsy from that patient re-
`vealed only chronic inflammatory change. Three
`other patients receiving copolymer 1 and one pa-
`tient receiving placebo withdrew because of tran-
`sient self-limited systemic reactions that were brief
`and not considered serious.
`M S relapse rates. During the 2-year trial, the
`copolymer 1-treated patients had 161 confirmed
`relapses and the placebo group had 210 confirmed
`relapses (table 4). The mean relapse rate (2 years)
`was 1.19 in the copolymer 1 group and 1.68 in the
`placebo group, a 29% reduction, which was statis-
`tically significant at the p = 0.007 level. Annual-
`ized relapse rates were 0.59 for the copolymer 1
`group and 0.84 for those receiving placebo. The
`median time to first relapse from baseline for the
`copolymer 1 group was 287 days and for the pla-
`cebo group it was 198 days, a difference that ap-
`proached statistical significance ( p = 0.097).
`Forty-two patients receiving copolymer 1 (33.6%)
`and 34 placebo patients (27.0%) were relapse-free
`throughout the trial ( p = 0.098). This result also
`approached statistical significance. When the re-
`lapse data were summarized in relation to base-
`line EDSS scores, it was found that patients with
`greater disability at entry had more relapses dur-
`ing the trial (figure 1). However, the therapeutic
`effect appeared to be most pronounced in patients
`with the lowest EDSS scores at entry (0 to 2), in
`
`Table 3. Patient exposure and duration of treatment
`
`I Duration of
`
`treatment
`
`<3
`>3-6
`>6-9
`>9-12
`>12-15
`>15-18
`>18-21
`>21-24
`224
`
`I
`
`Copolymer 1 (n = 125)
`
`n
`
`3
`3
`2
`5
`2
`2
`1
`1
`106
`125
`
`%
`
`2.4
`2.4
`1.6
`4
`1.6
`1.6
`0.8
`0.8
`84.8
`100
`
`Total
`patient
`months
`
`5.6
`13.6
`13.9
`49.4
`27.0
`33.1
`18.9
`21.3
`2,376.0
`2,725.3
`
`Placebo (n = 126)
`
`n
`
`4
`3
`0
`3
`3
`2
`1
`1
`109
`126
`
`%
`
`3.2
`2.4
`0.0
`2.4
`2.4
`1.6
`0.8
`0.8
`86.5
`100
`
`Total
`patient
`months
`
`3.8
`13.6
`0
`31.6
`41.2
`31.4
`20.5
`23.5
`2,615.9
`2,781.5
`
`July 1995 NEUROLOGY 45 1271
`
`Page 4 of 9
`
`YEDA EXHIBIT NO. 2004
`MYLAN PHARM. v YEDA
`IPR2017-00195
`
`

`

`Table 4. Relapse experience of copolymer 1 and placebo groups
`
`Copolymer 1
`(n = 125)
`
`Placebo
`(n = 126)
`
`Reduction
`vs placebo
`
`p Value
`
`1.68
`
`0.84
`210
`
`27.0%
`198
`
`34
`55
`37
`
`-29%
`
`0.007
`
`0.098
`0.097
`
`0.023
`
`Table 5. Disability experience measured by EDSS
`and ambulation index of copolymer 1 and placebo
`groups
`
`Copolymer 1 Placebo
`
`p Value
`
`Proportion of patients
`with a change in
`disability between
`baseline and conclusion
`Improved
`(EDSS decrease 21)
`No change
`Worse
`(EDSS increase 21)
`EDSS change from
`baseline (mean i SD)
`Proportion of
`progression-free
`patients
`Ambulation index
`(mean i SD)
`
`24.8%
`
`54.4%
`20.8%
`
`15.2%
`
`56.0%
`28.8%
`
`0.037*
`
`-0.05 f 1.13
`
`0.21 f 0.99
`
`0.023t
`
`78.4%
`
`75.4%
`
`NS
`
`0.27 f 0.94
`
`0.28 f 0.93
`
`NS
`
`EDSS Expanded Disability Status Scale.
`NS Not significantly different.
`* Categorical repeated measures.
`t Repeated-measures analysis of covariance.
`
`ceiving placebo, 75.4% showed no progression (NS).
`The mean ambulation index scores were also simi-
`lar between groups, 0.27 for copolymer 1-treated
`patients and 0.28 for those on placebo (NS).
`Adverse events. No clinically significant differences
`in vital signs were noted during the trial. The most
`commonly recognized adverse event was a localized
`injection-site reaction consisting of mild erythema
`and induration, which sometimes persisted for sev-
`eral days (table 6). It was observed at least once dur-
`ing 730 days of treatment in 90% of the copolymer 1-
`treated patients and in 59% of the patients receiving
`placebo. The other adverse event clearly related to
`therapy was a transient self-limited systemic reac-
`tion (table 7), which also was recognized in earlier
`
`copolymer 1 s t u d i e ~ . ~ . ~ This reaction was sporadic
`and unpredictable, occurred within minutes of an in-
`jection, and was characterized by a variable combina-
`
`'
`
`1.19
`
`0.59
`161
`
`33.6%
`287
`
`42
`60
`23
`
`Primary end points
`Relapse rate over 24 mo
`(covariate adjusted mean)
`Annualized relapse rate
`Observed relapses over 24 mo
`Secondary end points
`Proportion of relapse-free patients
`Median time to first relapse (days)
`Number of relapses per patient
`0
`1-2
`23
`
`2.5 -,
`4 I Copolymer 1 0 Placebo - 1*85 1.76
`h g 2.0
`
`1.44
`
`1.44
`
`c
`
`V."
`
`0-2
`
`2-4
`Baseline EDSS
`
`>4
`
`I
`
`Figure 1. Changes in relapse rate observed over 2 years,
`by baseline EDSS score. The numbers above each bar
`represent the mean 2-year relapse rate for each group.
`
`whom there was a 33% difference in the relapse
`rate in favor of copolymer 1.
`Neurologic disability. The effect of copolymer 1
`therapy on neurologic disability was evaluated in a
`series of secondary end points (table 5) based on
`the EDSS and ambulation index, and determined
`every 3 months by the examining neurologist. Fig-
`ure 2 shows that more patients receiving copolymer
`1 were improved whereas more patients on placebo
`were worse by one or more EDSS steps when com-
`pared between baseline and 24 months. This find-
`ing was statistically significant in favor of copoly-
`mer 1 for both the categorical repeated-measures
`analysis (p = 0.037) and the analysis from baseline
`to 24 months ( p = 0.024). The repeated-measures
`analysis of mean change in EDSS also significantly
`favored copolymer 1 ( p = 0.023). When progression
`to sustained disability was defined as an increase
`of one or more EDSS steps maintained for more
`than 90 days-that
`is, for two consecutive sched-
`uled visits-little
`difference was noted between
`groups. Of those patients treated with copolymer 1,
`78.4% were free of progression, while of those re-
`1272 NEUROLOGY 45 July 1996
`
`Page 5 of 9
`
`YEDA EXHIBIT NO. 2004
`MYLAN PHARM. v YEDA
`IPR2017-00195
`
`

`

`6o 1
`
`I Copolymer 1 0 Placebo
`54.4 56.0
`
`Table 7. Incidence of transient self-limited
`systemic reactions
`
`' Systemic reaction
`
`Primary symptoms
`Flushing without chest pain
`Chest pain without flushing
`Both chest pain and flushing
`Secondary symptoms
`Palpitation
`Anxiety
`Dyspnea
`
`"
`
`Improved No Change Worse
`Change in EDSS 21
`
`I
`
`Copolymer 1 Placebo
`(n = 125)
`(n = 126)
`n % n %
`
`19
`
`15.2
`
`6
`6
`7
`
`6
`2
`16
`
`3.2
`
`4
`
`2
`2
`0
`
`0
`2
`2
`
`Figure 2. Percent of patients who improved, were
`unchanged, or were worse by one or more EDSS steps
`between baseline and the last (24-month) measurement
`(repeated-measures ANCOVA). The numbers above the
`bars represent the percent of patients in the respective
`copolymer 1 or placebo group.
`
`Table 8. Number of episodes of transient self-
`limited systemic reactions experienced per patient
`over 2 years
`
`Copolymer 1
`(n = 125)
`n
`%
`
`Placebo
`(n = 126)
`n
`%
`
`Table 6. Observations on injection-site changes
`
`No. episodes*
`
`Copolymer 1
`n
`70
`
`Placebo
`9%
`
`n
`
`80
`71
`48
`34
`33
`27
`24
`
`64.00
`56.80
`38.40
`27.20
`26.40
`21.60
`19.20
`
`46
`16
`5
`8
`10
`45
`1
`
`36.51
`12.70
`3.97
`6.35
`7.94
`35.71
`0.79
`
`Pain
`Erythema
`Pruritus
`Inflammation
`Mass
`Ecchymosis
`Induration
`
`tion of flushing and chest tightness, accompanied at
`times by dyspnea, palpitations, or anxiety. It lasted
`between 30 seconds and 30 minutes, resolved sponta-
`neously without sequelae, and rarely was witnessed
`by medical personnel. It was reported at least once in
`15% of the copolymer 1-treated patients and in 3% of
`those receiving placebo, and was experienced seven
`times at most in any patient treated with copolymer
`1 and once in any patient receiving placebo (table 8).
`This reaction resulted in discontinuation of therapy
`by four patients in the copolymer 1 group and one in
`the placebo group. Other adverse events occurred ap-
`proximately equally in the copolymer 1- and placebo-
`treated groups.
`Although not an adverse event, pregnancy oc-
`curred in three women during the course of the trial,
`all in the copolymer 1-treated group. One elected to
`have a therapeutic abortion and continue, while two
`withdrew from the trial and delivered normal infants.
`Studies of blood and urine for common metabolic
`changes or hematologic abnormalities showed no dif-
`ferences between groups either at baseline or during
`the trial. ECGs at baseline and at the conclusion of
`the study were unchanged in both groups.
`
`10
`1
`4
`2
`3
`3
`1
`4
`7
`1
`* Over an average of 680 injections.
`
`8.0
`3.2
`2.4
`0.8
`0.8
`
`4
`0
`0
`0
`0
`
`3.2
`0
`0
`0
`0
`
`Discussion. This large multicenter trial success-
`fully confirmed the findings of an earlier pilot trial8
`showing that daily S.C. injections of 20 mg of copoly-
`mer 1 significantly reduced the relapse rate in re-
`lapsing-remitting MS patients. In addition, re-
`peated-measures analysis of the mean EDSS scores
`showed significant differences in disability between
`the treatment groups in favor of those receiving
`copolymer 1. Finally, the benign patient tolerance
`profile of earlier trials was maintained.
`The difference in mean relapse rate was the pri-
`mary end point in this 2-year study. Very few re-
`lapses were not confirmed by the examining neu-
`rologist within 7 days of onset of symptoms (as
`mandated in the protocol), so we believe this is a
`true picture of the clinical course experienced by
`these two well-matched groups. The difference in
`mean relapse rate was highly significant ( p =
`0.007). This clinical effect persisted through each 6-
`month interval of the study. The observations on
`the median number of days to first relapse and the
`proportion of relapse-free patients, although not
`statistically significant, did show strong trends in
`favor of copolymer 1 therapy.
`Figure 1 shows that patients with low EDSS
`scores at baseline were more likely to have had
`fewer relapses during the trial. A similar finding
`was evident in the copolymer 1 pilot study.8 Of in-
`July 1995 NEUROLOGY 45 1273
`
`Page 6 of 9
`
`YEDA EXHIBIT NO. 2004
`MYLAN PHARM. v YEDA
`IPR2017-00195
`
`

`

`terest, there appeared to be a correlation between
`EDSS at baseline and the subsequent relapse expe-
`rience (figure l). Patients with higher EDSS scores
`at entry may have had more active or virulent MS,
`showing not only more disability at baseline but
`also continued higher relapse activity during the
`course of the trial. This suggests that any large MS
`cohort is rather heterogeneous and that improved
`methods of patient classification must be found to
`aid in the design of future MS therapy trials.
`The difference in the mean relapse rate between
`groups in this study, although highly significant,
`was less pronounced than in the earlier copolymer
`1 pilot study.s The reason for this is unknown, but
`one possible reason may be the obvious difference
`in the patient populations studied. In this investi-
`gation, patients had a lower pre-study frequency of
`relapses and there were proportionally fewer pa-
`tients at the low end of the EDSS scale. One could
`argue that the cohort for this trial was more repre-
`sentative of the majority of relapsing-remitting MS
`populations.
`Now that both copolymer 1 and IFNB-lb2 have
`been shown to positively influence the relapse rate
`in relapsing-remitting MS, it is tempting to com-
`pare the magnitude of effect. The difference be-
`tween the high-dose IFNB-lb group and a placebo
`group was highly significant at the 0.0001 level.
`However, the annual relapse rate for IFNB-lb was
`0.84 whereas in this copolymer 1 study it was 0.59.
`The IFNB-lb high-dose group and the copolymer 1
`groups were of similar size (IFNB-lb = 115 and
`copolymer 1 = 1241, yet during 2 years of observa-
`tions, those receiving IFNB-lb experienced 173 re-
`lapses whereas the copolymer l-treated group ex-
`perienced only 161 relapses. Are such differences
`due to a different therapeutic effect or to inequali-
`ties in the populations selected for study? Probably
`only improved information on the natural history of
`MS, improved protocol design, and comparison of
`other measures of effect in future studies will an-
`swer this question.
`A positive influence on neurologic disability was
`suggested in earlier copolymer 1 clinical studies
`where there were encouraging trends but no signif-
`icant difference^.^,^ In the current investigation,
`several methods of analysis, based on the EDSS,
`showed that copolymer 1