ORIGINAL ARTICLES
`
`European/Canadian Multicenter, Double-
`Blind, Randomized, Placebo-Controlled
`Study of the Effects of Glatiramer Acetate
`on Magnetic Resonance Imaging–Measured
`Disease Activity and Burden in Patients with
`Relapsing Multiple Sclerosis
`
`Giancarlo Comi, MD,1 Massimo Filippi, MD,2 Jerry S. Wolinsky, MD,3 and the
`European/Canadian Glatiramer Acetate Study Group
`
`Two prior double-blind, placebo-controlled, randomized trials demonstrated that glatiramer acetate (GA) reduces relapse
`rates in patients with relapsing remitting multiple sclerosis (RRMS). This study was designed to determine the effect,
`onset, and durability of any effect of GA on disease activity monitored with magnetic resonance imaging (MRI) in
`patients with RRMS. Two hundred thirty-nine eligible patients were randomized to receive either 20 mg GA (n 5 119)
`or placebo (n 5 120) by daily subcutaneous injection. Eligibility required one or more relapses in the 2 years before
`entry and at least one enhancing lesion on a screening MRI. The study was a randomized, double-blind, placebo-
`controlled phase during which all patients studied underwent monthly MRI scans and clinical assessments over 9
`months. The primary outcome measure was the total number of enhancing lesions on T1-weighted images. Secondary
`outcome measures included the proportion of patients with enhancing lesions, the number of new enhancing lesions and
`change in their volume; the number of new lesions detected on T2-weighted images and change in their volume, and the
`change in volume of hypointense lesions seen on unenhanced T1-weighted images. Clinical measures of disease activity
`were also evaluated. The active treatment and placebo groups were comparable at entry for all demographic, clinical, and
`MRI variables. Treatment with GA showed a significant reduction in the total number of enhancing lesions compared
`with placebo (210.8, 95% confidence interval 218.0 to 23.7; p 5 0.003). Consistent differences favoring treatment
`with GA were seen for almost all secondary end points examined: number of new enhancing lesions ( p < 0.003),
`monthly change in the volume of enhancing lesions ( p 5 0.01), and change in volume ( p 5 0.006) and number of new
`lesions seen on T2-weighted images ( p < 0.003). The relapse rate was also significantly reduced by 33% for GA-treated
`patients ( p 5 0.012). All effects increased over time. Glatiramer acetate significantly reduced MRI-measured disease
`activity and burden.
`
`Ann Neurol 2001;49:290 –297
`
`Multiple sclerosis (MS) is an inflammatory demyelinat-
`ing disease of the central nervous system (CNS) affect-
`ing about 1 million people worldwide. The disease
`leads to substantial disability in most patients.1 Con-
`verging evidence supports the concept that MS is an
`immune-mediated disease of genetically susceptible
`subjects that is unleashed by one or more environmen-
`tal agents. A myelin antigen-specific, Th1-type T-cell–
`orchestrated immunopathogenesis directed at one or
`
`more immunodominant antigenic fragments of myelin,
`including myelin basic protein (MBP), proteolipid pro-
`tein (PLP),
`and/or
`oligodendrocyte
`glycoprotein
`(MOG), is postulated to be central to the immune dys-
`function that underlies the disorder.2
`Prior therapeutic strategies to modify the course of
`MS, based on nonselective immunosuppression to re-
`duce or eliminate the offending T cells, have been of
`limited benefit and are often associated with significant
`
`From the 1Clinical Trials Unit and 2Neuroimaging Research Unit,
`Department of Neuroscience, Scientific Institute Ospedale San
`Raffaele, University of Milan, Milan, Italy; and 3the University of
`Texas–Houston, Health Science Center, Houston, TX.
`Received Jul 17, 2000, and in revised form Sep 6. Accepted for
`publication Sep 26, 2000.
`
`290 © 2001 Wiley-Liss, Inc.
`
`Address correspondence to Dr Comi, Department of Neuroscience,
`Scientific Institute Ospedale San Raffaele, University of Milan, Via
`Olgettina 60, 20132 Milan, Italy. E-mail: g.comi@hsr.it
`
`MYLAN PHARMS. INC. EXHIBIT 1031 PAGE 1
`
`

`
`side effects. Recently, several classes of immunomodu-
`latory agents have been evaluated,3 and several formu-
`lations of recombinant interferon-b are now approved
`for the treatment of relapsing-remitting (RR) MS. The
`only other immunomodulator currently approved in
`several countries for the treatment of RRMS is glati-
`ramer acetate (GA) (Copaxonet, formerly known as
`copolymer 1).4 – 6
`Glatiramer acetate was developed for human use
`based on its effectiveness in preventing or reducing the
`severity of neurological disease in experimental allergic
`(EAE).7 Previous
`clinical
`trials
`encephalomyelitis
`showed that GA reduces relapse rate and the accumu-
`lation of disability in patients with RRMS.4 – 6 Al-
`though perivascular inflammation is a prominent ele-
`ment of the pathology of MS lesions that can be
`monitored using enhanced magnetic resonance imaging
`(MRI),8,9 limited data are available to assess the effect
`of GA on this and other MRI-defined aspects of the
`disease.10,11
`The present study was a placebo-controlled, double-
`blind clinical trial to determine the effect of GA on the
`cumulative number of enhancing lesions found on
`monthly MRI, and to assess other MRI measures that
`might be affected by treatment in a large cohort of pa-
`tients with RRMS. This study was designed to deter-
`mine whether treatment with GA is associated with a
`measurable effect on the inflammatory aspect of the
`disease, and to define the time course of the evolution
`of any effect.
`
`Patients and Methods
`Patients
`Twenty-nine centers in six European countries and Canada
`participated in the trial. Enrollment started in February 1997
`and concluded in November 1997. Of 485 patients screened
`with clinically definite MS,12 252 met all entry criteria and
`239 of these were randomized to treatment with placebo or
`GA, 20 mg daily by subcutaneous injection. Thirteen pa-
`tients were excluded at the entry visit based on the occur-
`rence of a relapse, steroid treatment, or abnormalities found
`on one or more screening tests. The clinical entry and exclu-
`sion criteria were nearly identical to those in the pivotal
`American trial.5 All patients had to have an RR course, a
`diagnosis of MS for at least 1 year, an age of 18 to 50 years
`inclusive,
`a Kurtzke Expanded Disability Status Scale
`(EDSS)13 score of 0 to 5, at least one documented relapse in
`the preceding 2 years, and at least one enhancing lesion on
`their screening brain MRI. Of note, prior controlled pivotal
`trials of GA in RRMS did not use MRI and required a min-
`imum of two clinical attacks in the 2 years before enroll-
`ment.4 – 6 Patients had to be clinically relapse-free and with-
`out steroid treatment in the 30 days before their pre-entry
`MRI. The previous use of GA or oral myelin led to exclu-
`sion, as did prior lymphoid irradiation, the use of immuno-
`suppressant or cytotoxic agents in the past 2 years, or the use
`of azathioprine, cyclosporine, interferons, deoxyspergualine,
`
`or chronic corticosteroids during the previous 6 months.
`Subjects receiving concomitant therapy with an experimental
`drug for MS or for another disease were ineligible. Patients
`with other serious intercurrent systemic or psychiatric ill-
`nesses, and those who were pregnant or unwilling to practice
`reliable methods of contraception during the course of the
`study were not enrolled. Patients with known hypersensitiv-
`ity to gadolinium-DTPA (Gd) or those unable to undergo
`repeated MRI studies were excluded. The ethical committees
`of all participating centers approved the study.
`
`Design
`The study was a double-blind, placebo-controlled, random-
`ized study lasting 9 months. For trial purposes a month was
`defined as 4 weeks (28 6 7 days). At the pre-enrollment visit
`eligible patients were informed about all aspects of the study
`and gave written informed consent. Eligible patients under-
`went physical and neurological examination, including EDSS
`and Ambulation Index,14 laboratory studies, and brain MRI.
`The coordinating center reviewed the results of
`the pre-
`enrollment evaluations including the brain MRI, and when all
`inclusion and exclusion criteria were satisfied, gave the ap-
`proval for patient enrollment. Patients were randomized to re-
`ceive 20 mg GA or placebo daily by subcutaneous injection.
`The randomization list, stratified by individual study cen-
`ter, was computer-generated by the TEVA Statistical Data
`Management Department. Equal allocation of the two treat-
`ment groups was used. At each study site a treating neurol-
`ogist was responsible for the overall medical management of
`the patient including safety monitoring. Vital signs and ad-
`verse effects were assessed monthly. An examining neurolo-
`gist was responsible for all scheduled neurological examina-
`tions
`and
`exacerbation follow-up. All
`patients
`had
`neurological evaluation every 3 months; additional assess-
`ments were carried out for symptoms suggestive of a relapse.
`MRI was performed on all patients at monthly intervals.
`Safety evaluations that included vital signs, hematology, and
`biochemical tests were performed every 3 months at all reg-
`ularly scheduled clinical visits. All personnel involved in the
`study were unaware of the treatment allocation. Patient and
`physician blinding were not formally assessed because the
`primary and secondary outcome measures were MRI param-
`eters. Nevertheless, both the treating neurologist and the pa-
`tient were informed of the importance of not discussing
`safety issues with the examining neurologist.
`A relapse was defined as the appearance of one or more
`new neurological symptoms, or the reappearance of one or
`more previously experienced ones. Patients were instructed to
`telephone their local center immediately if they perceived
`that they might be experiencing a relapse. A visit was ar-
`ranged within 7 days of notification. Neurological deteriora-
`tion had to last at least 48 hours and be preceded by a rel-
`atively stable or improving neurological state in the prior 30
`days. An event was counted as a relapse only when the pa-
`tient’s symptoms were accompanied by objective changes in
`the neurological examination corresponding to an increase of
`at least 0.5 points on the EDSS, or one grade in the score of
`two or more Functional Systems (FS), or two grades in one
`FS. Deterioration associated with fever or infection that can
`cause transient, secondary impairment of neurological func-
`
`Comi et al: Glatiramer Acetate and Brain-Enhanced MRI
`
`291
`
`MYLAN PHARMS. INC. EXHIBIT 1031 PAGE 2
`
`

`
`tion in MS patients were not considered relapses. Nor was a
`change in bowel, bladder, or cognitive function alone ac-
`cepted as a relapse. The principal investigator reviewed all
`exacerbation reports to check their consistency with this re-
`lapse definition. Relapses could be treated with a standard
`dose of 1.0 g intravenous methylprednisolone for 3 consec-
`utive days.
`
`Outcome Measures
`The primary outcome measure was the total number of en-
`hancing lesions. Secondary outcome measures included the
`total volume of enhancing lesions, proportion of patients
`with enhancing lesions, number of new enhancing lesions,
`number of new lesions on T2-weighted images, and percent
`change of lesion volume on T2-weighted images. The change
`in the volume of hypointense lesions on T1-weighted images
`was the final prespecified secondary outcome measure. Relapse
`rate and other clinical outcomes were tertiary outcomes.
`
`MRI Scanning and Analysis
`Before any participating center could enter patients into the
`trial, they were required to image a volunteer patient with
`clinically definite MS according to strict study imaging pro-
`tocol. The images were sent to the Neuroimaging Research
`Unit in Milan as film and electronic data for review to en-
`sure that high-quality imaging could be performed at each
`contributing center. Twenty-eight MRI sites were approved
`(21 scanners were operating at 1.5 T, five at 1.0 T, and two
`at
`0.5 T). Conventional
`spin-echo
`sequences
`(TR
`2200 –2800, TE 20 –50/60 –100, 3-mm slice thickness, and
`44 contiguous axial slices) were used to obtain proton den-
`sity and T2-weighted images. Two series of T1-weighted im-
`ages (TR 450 – 650, TE 10 –20, 3-mm slice thickness, and
`44 contiguous axial slices) were obtained before and 5 min-
`utes after the injection, through a long intravenous catheter,
`of 0.1 mmol/kg of Gd. The slices were positioned to run
`parallel to a line that joined the most infero-anterior and
`infero-posterior parts of the corpus callosum. At follow-up,
`patients were carefully repositioned following published
`guidelines.15
`Image quality was reviewed centrally according to prede-
`termined criteria. Unsatisfactory images were rejected, but
`not repeated except for the 36 weeks scan because of the
`frequent scanning interval. Patient MRI-based eligibility for
`study entry was determined by the central
`image analysis
`center. Identification of enhancing lesions, high signal inten-
`sity lesions on T2-weighted images, and hypointense lesions
`on T1-weighted unenhanced images was done by consensus
`of two experienced observers, as previously described.16,17
`Trained technicians then outlined the lesions using a semi-
`automated segmentation technique based on local threshold-
`ing,18 with reference to the marked hard copies. The lesion
`volumes were calculated automatically. In a previous study
`using the same measurement strategy,18 we showed that the
`median intraobserver coefficients of variation were 1.6%
`(range 5 1.8% to 6.2%) for T2 and 4.0% (range 5 2.2% to
`8.4%) for T1 lesion load.
`
`Statistical Analysis
`Analyses were based on an intention-to-treat data set. The last
`observation carried forward (LOCF) method was implemented
`
`292 Annals of Neurology Vol 49 No 3 March 2001
`
`to account for early discontinuation and missing data. The
`assessment of data as observed, without carrying forward miss-
`ing data (AS IS), was also planned to reduce any source of bias
`in the results due to early withdrawals or missing data. A
`baseline-adjusted analysis of covariance (ANCOVA) compared
`the two study arms for the primary end point, incorporating
`terms for treatment and center as main effects. Covariates were
`age, gender, baseline EDSS, disease duration, number of re-
`lapses in the 2 years before enrollment, and number of en-
`hancing lesions in the prerandomization scan. To better un-
`derstand the results, analyses were also performed using log
`transformation, rank transformation with ANCOVA, and
`quasi-likelihood Poisson regression. Tests for the comparability
`of study groups at baseline evaluated demographic, clinical,
`and MRI data. The continuous variables were examined using
`the two-sample two-sided t test or the Mann–Whitney test
`when appropriate. The categorical variables were checked for
`differences between groups using the x2 test or Fisher’s exact
`test as appropriate. All p values given are two tailed. Two in-
`terim analyses were planned when at least 65 and 130 patients
`had completed the study. For the first and second interim
`analyses, treatment effect was considered significant with a p
`value of #0.0005 and #0.014, respectively. The final analysis
`required a p value of 0.045 or better. Sample size was pro-
`jected based on literature data and on simulations modeled
`using a Poisson cyclic variable. We estimated that for an ex-
`pected treatment effect of 30% or more, a 9-month study
`with 85 patients in each arm (assuming a 20% dropout rate)
`would provide more than 85% power to detect a significant
`difference in the total number of enhancing lesions.
`
`Results
`Demographic and Baseline Characteristics
`Of the 239 enrolled patients, 119 were randomized to
`GA and 120 to placebo. Baseline demographic and
`clinical characteristics did not differ significantly be-
`tween the two study arms. They were also similar to
`those of patients who participated in the U.S. pivotal
`trial of GA and most other clinical trials in RRMS
`(Table 1). Baseline MRI characteristics were also not
`different between the two groups.
`
`Study Compliance
`Seven patients dropped out in each arm. Seven patients
`dropped out in the first trimester, five in the second
`trimester, and two in the third trimester. Two subjects
`in the placebo group and three in the GA group dis-
`continued treatment because of adverse experiences.
`One patient in the placebo arm discontinued treatment
`that he considered ineffective, another left because of
`poor compliance, one was lost to follow-up, and two
`refused to continue MRI monitoring. One subject dis-
`continued GA treatment when he moved away from the
`center, and another after a severe exacerbation. Four GA
`subjects withdrew their consent without providing a rea-
`son. Of 1309 planned MRI sessions in the GA group,
`1237 (94.5%) were available for analysis. The compara-
`ble proportion in the placebo arm was 96.3%.
`
`MYLAN PHARMS. INC. EXHIBIT 1031 PAGE 3
`
`

`
`Table 1. Baseline Clinical and MRI Characteristics (Mean 6 SD) of Trial Subjects and Comparison with Clinical Features of
`U.S. Pivotal Trial Cohorts
`
`European/Canadian
`
`U.S. Pivotal
`
`N
`Age (years)
`Disease duration (years)
`Prior 2-year relapse rate
`EDSS
`Ambulation index
`Enhancing lesion number
`New enhancing lesion numbera
`Enhancing lesion volume (ml)
`New T2 lesion numbera
`T2 lesion volume (ml)
`T1 hypointense lesion volume (ml)
`
`Placebo
`
`120
`34.0 6 7.5
`8.3 6 5.5
`2.5 6 1.4
`2.4 6 1.2
`1.2 6 1.1
`4.4 6 7.1
`2.6 6 4.1
`0.7 6 2.2
`1.2 6 1.7
`20.5 6 18.8
`4.0 6 4.9
`
`GA
`
`119
`34.1 6 7.4
`7.9 6 5.5
`2.8 6 1.8
`2.3 6 1.1
`1.1 6 0.9
`4.2 6 4.8
`2.5 6 3.5
`0.6 6 0.7
`1.0 6 1.5
`20.0 6 17.2
`3.4 6 3.9
`
`Placebo
`
`126
`34.3 6 6.5
`6.6 6 5.1
`2.9 6 1.1
`2.4 6 1.3
`1.1 6 0.9
`
`GA
`
`125
`34.6 6 6.0
`7.3 6 4.9
`2.9 6 1.3
`2.8 6 1.2
`1.2 6 1.0
`
`aNew enhancing and T2 lesions at entry were determined by comparing the pre-enrollment scans (obtained to assess whether patients met the
`MRI criteria for study entry) and the baseline scans. According to protocol, the maximum elapsed interval between these two scans was 28 days.
`On average, it was 22.2 days for placebo patients and 21.0 days for GA patients.
`
`MRI Outcomes
`The mean total number of enhancing lesions was
`36.80 for the placebo group and 25.96 for the GA
`group (Fig 1). The mean reduction in the total number
`of enhancing lesions in the GA group compared with
`the placebo group was 210.8 (95% confidence interval
`[CI], 218.0 to 23.7; p 5 0.003), a 29% reduction.
`When analyzed AS IS, the results were similar, reflect-
`ing the limited loss of MRI data, with a 35% reduction
`in the total number of enhancing lesions for the GA
`treated arm compared with the placebo group (33.7
`versus 21.8; p , 0.001). The mean number of enhanc-
`
`Fig 1. Mean number of cumulative enhancing lesions identi-
`fied on all postgadolinium T1-weighted images over the 9
`months of the study. Data are shown using the last available
`observation carried forward (LOCF) when MRI data were
`missing for any specific time interval for a given patient. They
`are also shown using all available data (As Is) without an
`adjustment for missing values.
`
`ing lesions per patient per month differed between
`treatment groups in favor of GA (2.9 versus 4.1; p ,
`0.005). Repeated measures analysis showed a drug in-
`teraction with time (p 5 0.027). Statistically significant
`differences in the mean number of enhancing lesion
`per patient per month emerged between the groups at
`month 6 (Fig 2). The magnitude of the treatment ef-
`fect seen on the total number of new enhancing lesions
`paralleled that observed for the total number of en-
`hanced lesions with a 33% reduction for the GA treat-
`ment group (17.4 versus 26.0; p , 0.003). The time
`course of the effect of GA on the number of new en-
`hancing lesions was also similar to that on the number
`of enhancing lesions (data not shown).
`The results of secondary MRI end points are sum-
`
`Fig 2. Mean number of total enhancing lesions per subject
`observed at each month on study using the last observation
`carried forward approach. Repeated measures analysis favored
`a treatment effect for glatiramer acetate ( p 5 0.003). Statis-
`tically significant differences first emerged after 5 months of
`continued therapy from randomization.
`
`Comi et al: Glatiramer Acetate and Brain-Enhanced MRI
`
`293
`
`MYLAN PHARMS. INC. EXHIBIT 1031 PAGE 4
`
`

`
`marized in Table 2. The percentage of patients with
`scans free from enhancing lesions at baseline were GA
`17.6% and placebo 18.3%. Only six patients in the
`placebo group and three in the GA treated group were
`inactive over the entire study. However, on monthly
`comparison, a positive effect of GA emerged during the
`third trimester (12.5% versus 27.7%; odds ratio [OR]
`3.86; CI 1.76 to 8.48; p , 0.001). During the entire
`study, the mean percentage of scans without any en-
`hancing lesions
`(i.e., “inactive” scans) was 28.7%
`(SD 5 2.8%) in the placebo group and 35.8% (SD 5
`2.9%) in the GA-treated group ( p 5 0.04). This dif-
`ference was not significant during the first trimester of
`the study (placebo: 29.4 6 3.4%, GA: 26.2 6 3.5%),
`but it became evident during the second (placebo:
`29.9 6 3.6%, GA: 39.8 6 3.8%, p 5 0.03) and the
`third (placebo: 26.0 6 3.8%, GA: 43.8 6 3.9%, p 5
`0.0002) trimesters. Repeated measures analysis on the
`cumulative monthly change from baseline in enhancing
`lesion volume showed a significant difference that fa-
`vored active treatment ( p , 0.01). Figure 3 shows the
`cumulative enhancing lesion volume on a monthly ba-
`sis for the entire study duration. The two curves first
`diverge at month 4 and the separation increased in the
`following months. The mean total number of new T2
`lesions was 13.5 in the placebo group and 9.4 in the
`GA group, a 30% difference ( p , 0.003). Differences
`in the accumulation of new T2 lesions over time in the
`two groups paralleled those observed for enhancing le-
`sions. By the second trimester, the rate of accumula-
`tion of new T2 lesions in the GA-treated group first
`slowed and then continued to diverge from the placebo
`group, becoming significant after month 6 (Fig 4). The
`median percentage change in T2 lesion volume from
`baseline to the end of the trial was 20.6% in the pla-
`cebo group and 12.3% in the GA group (a 40% re-
`duction; p 5 0.0011). Intergroup divergence in the ac-
`cumulation of T2 lesion volume was again evident
`during the second trimester and became significant in
`the third. The median change (baseline to termina-
`tion), in T1-weighted hypointense lesion volume was
`
`Table 2. Secondary MRI End Points
`
`Total no. new enhancing lesions
`Change from baseline to month 9 in enhancing lesion
`volume (ml)
`Total no. new T2 lesions
`Change from baseline to month 9 in T2 lesion volume (ml)
`Change from baseline to month 9 in hypointense T1 lesion
`volume (ml)
`
`Values presented are all unadjusted means.
`ap values refer to comparisons based on adjusted means.
`
`294 Annals of Neurology Vol 49 No 3 March 2001
`
`Fig 3. Cumulative median enhancing lesion volume from ran-
`domization, displayed in milliliters. Statistically significant
`differences emerged during the third trimester.
`
`266 ml for the GA arm and 425 ml for the placebo
`group. This 37% reduction in favor of active treatment
`was not significant.
`
`Clinical Outcomes
`The observed mean relapse rate was 33% lower in the
`GA group (0.51 relapses/subject) than in the placebo
`arm (0.76 relapses/subject). This difference was statis-
`tically significant ( p 5 0.012). The corresponding an-
`nualized relapse rates were 0.81 and 1.21. The num-
`bers of relapses were similar in the two arms over the
`first two trimesters. In contrast, during the third tri-
`mester only five relapses occurred in the GA group,
`whereas 26 relapses occurred in the placebo group. The
`proportion of
`relapse-free patients was
`slightly in-
`creased in the GA group (55.5% versus 49.2%, OR
`1.47, CI 0.84 to 2.56; p 5 0.175). However, the per-
`centage of patients with two or more relapses dropped
`from 15.8% in the placebo group to 6.7% in the GA
`group. A significant correlation between cumulative
`number of enhancing lesions and total number of re-
`lapses over the study period was observed in both pla-
`cebo (Spearman rank correlation coefficient [SRCC] 5
`0.35; p 5 0.0001)
`and in GA-treated patients
`(SRCC 5 0.24, p 5 0.01). Steroid courses were ad-
`
`Placebo
`
`Glatiramer Acetate
`
`Mean 6 SE Median Mean 6 SE
`
`Median
`
`p
`
`9.0 ,0.003a
`17.4 6 2.2
`26.0 6 3.1
`13.5
`2105.1 6 177.4 252.5 2245.3 6 70.8 2169
`0.01
`
`13.7 6 1.1
`4.7 6 0.9
`1.3 6 0.2
`
`8.0
`3.0
`0.4
`
`9.4 6 1.1
`3.0 6 0.4
`0.8 6 0.2
`
`5.0 ,0.003a
`1.7
`0.006
`0.3
`0.14
`
`MYLAN PHARMS. INC. EXHIBIT 1031 PAGE 5
`
`

`
`Few subjects in either study treatment arm experienced
`more than one of these reactions.
`
`Discussion
`This study was designed to evaluate the effect of GA
`on MRI-monitored features of the pathology of MS.
`As a consequence, although patients were selected to
`clinically resemble those studied in previous controlled
`trials of GA in RRMS,4 – 6 they were also required to
`have at least one enhancing lesion. This is why the fre-
`quency of active scans and the number of enhancing
`lesions found at entry exceeded those anticipated from
`studies of similar RRMS patient cohorts selected by clin-
`ical criteria alone.4 – 6 Clearly, this makes this cohort of
`patients difficult to compare with those of previous trials
`testing the effect of the same or different treatments.
`Limited spontaneous decline occurred in the high rate of
`subclinical disease activity seen in the placebo group over
`the subsequent 9 months of the trial. A recent meta-
`analysis study concluded that the rate of enhancement is
`correlated with relapse rate over the next few months in
`patients with RRMS or secondary progressive MS.20
`This may explain the higher relapse rate found in both
`arms in this study compared with the pivotal North
`American study.5
`A clear effect of therapy with GA emerged for al-
`most all MRI-monitored parameters
`studied. The
`number of enhancing lesions was reduced by 29%
`compared with placebo. This finding was supported by
`a reduction in the number of new enhancing lesions
`and the total volume of tissue that enhanced. The
`pathological correlates of enhancement are perivascular
`inflammation and regional blood brain barrier disrup-
`tion.21 Consistent with an effect on the occurrence,
`number, and extent of tissue enhancement, the number
`of new lesions found on T2-weighted images and the
`total volume of tissue involved on the T2-weighted
`scan were also reduced by active treatment with GA.
`The pathological substrate of the lesions seen on T2-
`weighted images is heterogeneous, but in aggregate is
`generally accepted as a measure of the lesion load or
`disease burden.8 Trends were seen that favored active
`treatment on the hypointense lesion volume that did
`not attain significance. T1-hypointense lesions reflect
`regions with severe tissue destruction.22
`The time of appearance of both clinical and MRI-
`monitored therapeutic effects in this trial is consistent
`with the apparent mechanism of action of GA. In vitro
`studies show that GA rapidly, avidly, and promiscu-
`ously binds to various major histocompatibility com-
`plex class II molecules and displaces other antigens, in-
`cluding putative pathogenic myelin autoantigens to
`prevent antigen specific T-cell activation.23,24 The im-
`portance of antigen displacement to the observed effect
`of the drug is uncertain. However, further processing
`of GA bound to antigen presenting cells is not neces-
`
`Comi et al: Glatiramer Acetate and Brain-Enhanced MRI
`
`295
`
`Fig 4. Cumulative mean number of new lesions observed on
`the T2-weighted images at each month on study. Statistically
`significant differences first emerged after 5 months of contin-
`ued therapy from randomization.
`
`ministered to 33.6% and 39.2% of the patients in the
`GA and placebo arms, respectively. There were 84
`steroid-treated relapses among the placebo group and
`54 steroid-treated relapses
`for
`those on GA. The
`relapse-related hospitalizations were 16 in the GA
`group and 30 in the placebo group. As expected for a
`short duration study, no important changes were ob-
`served between baseline and final EDSS scores in either
`group (mean EDSS change from baseline, 10.02 GA,
`10.05 placebo).
`
`Safety and Tolerability
`Consistent with all prior studies of GA, the drug was
`well tolerated in this study. No significant differences
`between the groups were found for vital signs and met-
`abolic or hematologic parameters. Serious adverse ex-
`periences were reported in 10 GA (8.4%) and six pla-
`cebo (5.0%) patients. These events were predominantly
`classified as unrelated to the study drug. Investigators
`regarded one report of palpitation and one report of
`pain in GA-treated subjects as probably and possibly
`drug related, respectively. As anticipated by earlier ex-
`perience,4 – 6,19 two types of adverse events were seen
`frequently: local injection site reactions and immediate
`postinjection systemic reactions (IPISR). At least one
`local injection site reaction was reported by 70.6% of
`those receiving GA and by 28.3% in the placebo arm.
`However, the skin reactions were mild in most of the
`cases and necrosis was never observed. The IPISR that
`consists of a variable combination of facial flushing,
`chest tightness, dyspnea, palpitation, tachycardia, and
`anxiety was significantly associated with GA. It was un-
`predictable, started seconds or a few minutes after the
`injection, lasted 10 to 30 minutes, and always resolved
`without
`sequelae. Considering any of
`the above-
`mentioned symptoms as significant of an IPISR, the
`incidence of patients reporting at least one episode was
`37.8% in the GA and 13.3% in the placebo groups.
`
`MYLAN PHARMS. INC. EXHIBIT 1031 PAGE 6
`
`

`
`sary for its efficient presentation to T cells.25 Poten-
`tially more relevant in vivo studies showed that over
`time GA treatment resulted in GA-specific T cells
`that on in vitro stimulation with GA proliferated
`and secreted anti-inflammatory cytokines,
`including
`interleukin-4, interleukin-10, and transforming growth
`factor-b.26 These are cytokine secretion profiles that
`typify T cells of the Th2 regulatory or suppressor cell
`type. A proportion of the GA-specific T cells can be
`cross-stimulated by MBP and its immunodominant
`fragments to secrete the same regulatory cytokines.
`These cells transfer protection from clinical disease in
`the EAE model. A longitudinal study performed in 10
`RRMS patients treated with GA demonstrated a reduc-
`tion in proinflammatory cytokines and an increase of
`anti-inflammatory cytokines.27 Anti-inflammatory cy-
`tokine levels peaked during the first 6 months of treat-
`ment and then gradually decreased, whereas proinflam-
`matory cytokine levels continued to decrease. The
`time-dependent effect of treatment with GA seen in
`the present trial is consistent with some delay in the
`induction and expansion of an appropriate protective
`GA-specific regulatory T-cell population. It is also in-
`teresting that the magnitude of the effect on the in-
`flammatory aspect of the disease is modest, possibly
`consistent with a major site of action of these cells
`within the CNS that limits the extent of tissue damage.
`The latter is reflected by the parallel effect on new le-
`sions seen on T2-weighted images and on the accumu-
`lated disease burden.
`This trial has shown that the effect of GA on clinical
`and most MRI measures of disease activity is compa-
`rable to that of the various formulations of interferon-
`b.28 –31 The exquisite sensitivity of MRI better refined
`the time course of the development of the effects of
`GA on both clinical and MRI-defined disease activity
`seen in this study. However, this delay is partly consis-
`tent with accumulating evidence of the time course of
`immunological changes that appear after initiation of
`GA treatment.2,32 In contrast, at most clinically effec-
`interferon-b therapy is associated with a
`tive doses,
`substantial reduction of enhancement that is rapid in
`onset, being established within weeks of initiation of
`therapy.33 This study was of 9-months’ duration, and
`as a consequence,
`it cannot answer whether, once
`achieved, the effect of GA on enhancement is main-
`tained. Additional
`information on this issue will be
`provided by the second part of this study, which is an
`additional 9-month open-label phase. The uncertain
`relationship between frequency of enhancement and
`accumulation of irreversible disability20 also does not
`allow us to comment on whether the ability of the
`drug to suppress inflammation is one of the fundamen-
`tal mechanisms through which GA has shown to re-
`duce the accumulation of disability in patients with
`RRMS.5,6
`
`296 Annals of Neurology Vol 49 No 3 March 2001
`
`The study confirms the excellent safety and tolera-
`bility profile of GA. No unanticipated side effects
`emerged. Of note were the IPISR and injection site
`reactions that were frequent, usua