Randomized, double-blind, dose(cid:173)
`comparison study of glatiramer acetate in
`relapsing-remitting MS
`
`J.A. Cohen, MD; M. Rovaris, MD; A.D. Goodman, MD; D. Ladkani, PhD; D. Wynn, MD;
`and M. Filippi, MD; for the 9006 Study Group
`
`Abstract-O~jective: To evaluate the safety, tolerability, and efficacy of glatiramer acetate (GA) 40 mg daily vs the
`approved 20-mg formulation in relapsing-remitting multiple sclerosis. Methods: Eligibility criteria included clinically
`definite multiple sclerosis, Expanded Disability Status Scale score 0 to 5.0, no previous use of GA, at least one relapse in
`the previous year, and 1 to 15 gadolinium-enhancing (GdE) lesions on a screening MRI. MRI was repeated at months 3, 7,
`8, and 9, and neurologic examinations were performed at baseline and months 3, 6, and 9. Results: Of 229 subjects
`screened, 90 were randomly assigned to GA 20 mg (n = 44) or 40 mg (n = 46). The groups were well matched at baseline
`for demographic, clinical, and MRI characteristics. The primary efficacy endpoint, total number of GdE lesions at months
`7, 8, and 9, showed a trend favoring the 40-mg group (38% relative reduction, p = 0.0898). A difference between the two
`dose groups emerged as early as month 3 (52% reduction; p = 0.0051). There was a trend favoring the 40-mg group for
`relapse rate with benefit on proportion of relapse-free subjects (p = 0.0183) and time to first relapse (p = 0.0367). GA 40
`mg was well tolerated, with an overall safety profile similar to that of 20 mg. Some features of injection site reactions and
`immediate postinjection reactions were more common and severe with the higher dose. Conclusions: Glatiramer acetate
`(GA) 40 mg was safe and well tolerated. The overall efficacy results suggested that a 40-mg dose of GA may be more
`effective than the currently approved 20-mg daily dose in reducing MRI activity and clinical relapses.
`
`NEUROLOGY 2007;68:939-944
`
`4 and a meta-analysis of those
`Three pivotal trials1
`-
`studies5 support the benefit of glatiramer acetate
`(GA) 20 mg by daily subcutaneous injection on re(cid:173)
`lapse rate, accumulation of disability, and MRI le(cid:173)
`sion activity
`in
`relapsing-remitting multiple
`sclerosis (RRMS). In addition, a long-term open-label
`study6 and use for more than a decade in clinical
`practice also demonstrate the safety and tolerability
`of GA.
`There are few published data regarding other
`doses of GA. In an early study/ three patients with
`acute disseminated encephalomyelitis were treated
`with GA 2 to 3 mg by daily IM injection for 2 weeks.
`In the same study, four patients with terminal mul(cid:173)
`tiple sclerosis (MS) were treated with 2 to 3 mg every
`2 to 3 days for three weeks and then 2 to 3 mg
`
`Additional material related to this article can be found on the Neurology
`Web site. Go to www.neurology.org and scroll down the Table of Con(cid:173)
`tents for the March 20 issue to find the title link for this article.
`
`weekly over 2 to 5 months. No adverse effects were
`seen in this small pilot study, but no definite conclu(cid:173)
`sions regarding efficacy could be drawn. In a second
`preliminary open-label study, 8 12 patients with
`chronic progressive MS and 4 with RRMS were
`treated with GA 5 mg by IM injection five times per
`week for 3 weeks, three times per week for 3 weeks,
`twice per week for 3 weeks, and then once per week
`for the remainder of a 6-month period. Many of the
`patients demonstrated initial improvement, but over
`time and as the dose was reduced, the response dis(cid:173)
`appeared. Over the next year, the dose was gradu(cid:173)
`ally increased to 20 mg per day. Among the 15
`subjects who completed the study, 2 of 3 patients
`with RRMS and 3 of 12 with chronic progressive MS
`were described as improved. GA was well tolerated
`in this study. The randomized controlled trial of 50
`subjects with RRMS by Bornstein et al. 1 tested 20
`mg by daily subcutaneous injection and demon(cid:173)
`strated a beneficial treatment effect on relapses and
`disability progression with good tolerability. In a
`
`From the Mellen Center for MS Treatment and Research (J.A.C.), Cleveland Clinic Foundation, Cleveland, OH; Neuroimaging Research Unit (M.R., M.F.),
`Department of Neurology, San Raffaele Scientific Institute, Milan, Italy; Department of Neurology (A.D.G.), University of Rochester, Rochester, NY; Teva
`Pharmaceuticals (D.L.). Petach Tiqva, Israel; and Consultants in Neurology (D.W.), Northbrook, IL.
`Disclosure: This study was funded by Teva Pharmaceuticals, of which Dr. Ladkani is an employee. The other authors have served as paid consultants or
`speakers for Teva Pharmaceuticals (less than $10,000 per year) and have received research support to their institutions, but none has a personal financial
`investment, ownership, equity. or other financial holdings. Preliminary results were presented at the annual meetings of the American Academy of
`Neurology in April 2006 in San Diego, CA, and the European Committee for Treatment and Research in Multiple Sclerosis in September 2006 in Madrid,
`Spain.
`Received June 7, 2006. Accepted in final form November 28, 2006.
`Address correspondence and reprint requests to Dr. Jeffrey A. Cohen, Mellen Center U-10, Cleveland Clinic Foundation, 9500 Euclid Avenue, Cleveland. OH
`44195; e-mail: cohenj@ccf.org
`
`Copyright © 2007 by AAN Enterprises, Inc. 939
`
`MYLAN PHARMS. INC. EXHIBIT 1056 PAGE 1
`
`

`

`subsequent study of 106 subjects with chronic pro(cid:173)
`gressive MS, 9 GA was administered subcutaneously
`at a dose of 15 mg twice per day for 2 years. This
`dose was well tolerated and demonstrated benefit on
`some disability endpoints. All subsequent studies of
`GA administered by injection used a 20-mg daily
`dose, the currently approved regimen. No dose com(cid:173)
`parison study has been published. In this study, we
`sought to evaluate the efficacy, tolerability, and
`safety of GA at a dose of 40 mg by daily subcutane(cid:173)
`ous injection in RRMS.
`
`Methods. Subjects. Eighteen centers in the United States par(cid:173)
`ticipated in this study (see appendix E-1 on the Neurology Web
`site at www.neurology.org for investigators and study commit(cid:173)
`tees). Enrollment started in October 2003 and was completed in
`January 2005. The inclusion and exclusion criteria were nearly
`identical to those in the European!Canadian MRI trial." Key inclu(cid:173)
`sion criteria included clinically definite MS, 10 age 18 to 50 years
`inclusive, Expanded Disability Status Scale (EDSS) score 0 to 5.0,
`at least one clinical relapse in the previous year, and 1 to 15
`gadolinium-enhancing (GdE) lesions on an MRI scan obtained at
`screening. Key exclusion criteria included relapse or steroid treat(cid:173)
`ment within 30 days of the screening visit or between the screen(cid:173)
`ing and baseline visits, previous GA therapy, interferon therapy
`within 60 days, immunosuppressant therapy within 6 months,
`previous use of cladribine or total lymphoid irradiation, investiga(cid:173)
`tional therapy within 6 months, known sensitivity to mannitol,
`and inability to undergo MRI with paramagnetic contrast agents.
`The protocol and consent documents were approved by the institu(cid:173)
`tional review boards of the participating centers. Subjects
`provided written informed consent before undergoing any study(cid:173)
`related procedures.
`Treatment. Eligible subjects were equally randomized to re(cid:173)
`ceive GA (Copaxone®) 20 mg or 40 mg by a single daily subcutane(cid:173)
`ous injection for 9 months. The randomization list, stratified by
`study center, was computer generated by the Teva Statistical
`Data Management Department. The drug preparations were iden(cid:173)
`tical except for GA concentration. Subjects and all personnel in(cid:173)
`volved in the study were blinded to treatment assignment. Subject
`and investigator blinding were not formally assessed.
`Design. The trial was a multicenter, randomized, double(cid:173)
`blind, parallel-group, dose-comparison study lasting 9 months. For
`trial purposes, a month was defined as 28 ± 4 days. At each study
`site, a treating neurologist was responsible for the overall medical
`management of subjects including safety monitoring. An examin(cid:173)
`ing neurologist performed a standardized neurologic examination,
`Timed 25-Foot Walk (T25FW), and calculated Functional System
`scores and EDSS score (Neurostatus, L. Kappos, MD, Department
`of Neurology, University Hospital, Basel, Switzerland) at sched(cid:173)
`uled and unscheduled visits.
`At the screening visit, potential subjects were informed about
`all aspects of the study, gave written informed consent, and then
`underwent physical and neurologic examinations including EDSS
`and T25FW, laboratory studies, and brain MRI. The MRI analysis
`center reviewed the results of the MRI and notified the site
`whether the subject qualified based on MRI criteria. Subjects re(cid:173)
`turned for the baseline visit within 14 days of the screening MRI
`and were randomized to one of two doses of GA using an interac(cid:173)
`tive voice response system. MRI obtained at screening also served
`as the pretreatment baseline MRI. MRI was repeated at months 3,
`7, 8, and 9 or at early termination (if the subject had been in the
`trial at least 3 months). Scheduled MRI scans were not delayed
`because of steroid treatment for a confirmed relapse. Neurologic
`evaluations were performed at baseline and then every 3 months.
`Vital signs (blood pressure, pulse, and temperature), adverse
`events, and concomitant medications were assessed at baseline
`and then at months 1, 3, 6, 7, 8, and 9 or early termination.
`Height and weight were measured at screening, and weight was
`measured at month 9 or early termination. Laboratory safety as(cid:173)
`sessments (hematology, serum chemistries, and urinalysis) and
`EKG were performed at baseline and then at months 1, 3, 6, and 9
`or early termination. Blood samples for anti-GA antibodies were
`
`940 NEUROLOGY 68 March 20, 2007
`
`collected at baseline and then at every 3 months or early
`termination.
`A relapse was defined as the appearance of one or more new
`neurologic symptoms, or the reappearance of one or more previ(cid:173)
`ously experienced symptoms lasting at least 48 hours, not accom(cid:173)
`panied by fever or infection, and preceded by a stable or improving
`neurologic state over the previous 30 days. Subjects were in(cid:173)
`structed to notify the study center of a potential change in neuro(cid:173)
`logic status immediately, and an unscheduled visit was conducted
`within 7 days of notification. An event was counted as a relapse
`only when the subject's symptoms were accompanied by objective
`changes in the examining neurologist's examination correspond(cid:173)
`ing to an increase of at least 0.5 steps on the EDSS, one grade in
`two or more Functional System scores, or two grades in one Func(cid:173)
`tional System score. Isolated changes in bowel, bladder, and cog(cid:173)
`nitive function did not qualify as relapses. The treating
`neurologist determined whether the change in symptoms qualified
`as an on-study relapse, which could be treated at the discretion of
`the treating neurologist with a standard 1,000-mg dose of IV
`methylprednisolone for 3 consecutive days without an oral taper.
`The Steering Committee supervised the conduct of the study.
`An independent Data Safety Monitoring Board met five times via
`teleconference during the trial to review study conduct and
`blinded safety data and a sixth time after completion of the trial
`to review unblinded safety results. They had the authority to
`recommend discontinuation of the trial for safety concerns.
`MRI scanning and analysis. The MRI Analysis Center in the
`Neuroimaging Research Unit, San Raffaele Scientific Institute,
`Milan, Italy, served as the MRI analysis center. Participating
`centers submitted a test scan of a volunteer with clinically definite
`MS for approval before enrolling subjects. All sites had 1.0- or
`1.5-T scanners. Dual-echo spin-echo sequences (repetition time
`[TR] 2,200 to 3,000, echo time (TE) 15 to 50/80 to 100, echo train
`length 4 to 6, 3-mm slice thickness, and 44 contiguous axial slices)
`were used to obtain proton density and T2-weighted images. T1-
`weighted images (TR 450 to 650, TE 10 to 20) with the same scan
`geometry were obtained 5 minutes after injection of 0.1 mmollkg
`of Gd. Slices were positioned to run parallel to a line joining the
`most inferoanterior and inferoposterior parts of the corpus callo(cid:173)
`sum. Subjects were carefully repositioned at follow-up according
`to published guidelines.ll
`Image quality was reviewed at the MRI analysis center using
`predetermined criteria. Unsatisfactory images were rejected, but
`not repeated. Identification of GdE, T2-hyperintense, and T1-
`hypointense lesions was performed by consensus of two experi(cid:173)
`enced observers, as previously described. 12,13 Trained technicians
`then outlined the lesions using a semiautomated segmentation
`technique based on local thresholding, 14 with reference to marked
`hard copies. Lesion volumes were calculated automatically. In a
`previous study using the same measurement technique, the me(cid:173)
`dian intraobserver coefficients of variation were 1.6% (range 1.8%
`to 6.2%) for T2 and 4.0% (range 2.2% to 8.4%) for T1lesion loads. 14
`Treating and examining neurologists at the sites were blinded to
`MRI results during the study.
`Outcome measures. The primary efficacy outcome measure
`was total number of GdE MRI lesions at months 7, 8, and 9.
`Secondary outcome measures included total number of new GdE
`lesions at months 8 and 9, total number of new T2-hyperintense
`lesions at months 8 and 9, change from baseline to termination in
`T2-hyperintense lesion volume, relapse rate, and change from
`baseline to each visit in the T25FW. Prespecified exploratory end(cid:173)
`points included change from baseline to termination in total GdE
`lesion volume, change from baseline to termination in total T1-
`hypointense lesion volume, MRI metrics at month 3, and change
`from baseline to each visit in EDSS. Post hoc analyses included
`time to first relapse, proportion of relapse-free subjects, and re(cid:173)
`sponder analyses,
`Statistical analysis. All efficacy and safety analyses were per(cid:173)
`formed on the intent-to-treat (ITT) cohort, defined as all random(cid:173)
`ized subjects. For analysis of the primary efficacy endpoint, the
`additional condition of having at least one MRI scan at month 7,
`8, or 9 was required to permit statistical analysis, resulting in 39
`subjects on GA 20 mg and 42 subjects on GA 40 mg. Analysis of
`the primary efficacy outcome and other analyses of GdE or new
`T2-hyperintense lesions used quasi-likelihood (overdispersed)
`Poisson regression (SAS PROC GENMOD) employing DIST =
`POI and DSCALE in the options of the MODEL statement with
`
`MYLAN PHARMS. INC. EXHIBIT 1056 PAGE 2
`
`

`

`Screen failures (n=139)
`· No enhancing lesions (135)
`• Withdrew consent (3)
`• Lost to follow-up (1)
`
`Screened
`(n=229)
`
`Randomized
`(n=90)
`
`GA20mg
`(n=44)
`
`GA40mg
`(n=46)
`
`Withdrew from study (n=6)
`o Lost to follow-up (2)
`o Withdrew consent (3)
`• Adverse event (1)
`
`Withdrew from study (n=6)
`o Lost to follow-up (2)
`• Adverse event (4)
`
`9-month follow-up on
`treatment
`(n=38)
`
`9-month follow-up on
`treatment
`(n=40)
`
`Figure 1. Trial profile. GA = glatiramer acetate.
`
`an offset based on the log of proportion of available scans (1!3, 2/3,
`or 3/3). Baseline GdE lesion count and center effects were used as
`covariates in the model in addition to the treatment effect. The
`numbers of subjects withdrawing early due to adverse events were
`compared using the Fisher exact test, and time to withdrawal was
`analyzed by log-rank test. Relapse rates were analyzed using Pois(cid:173)
`son regression with relapse rate in the year before entry and
`baseline EDSS as covariates. Analysis of proportion of relapse-free
`subjects used the x2 test. Time to first confirmed relapse was
`displayed graphically by Kaplan-Meier curves and analyzed by
`log-rank test. In post hoc analyses, subjects were classified as
`responders or nonresponders based on the occurrence of relapses,
`EDSS progression, and the presence of GdE or new T2-
`hyperintense lesions at months 7, 8, and 9. Analyses based on two
`definitions were performed using logistic regression adjusted for
`site and baseline GdE lesion number. All reported p values were
`two-tailed. Data analysis was performed by Teva Pharmaceutical
`Industries, Israel. The authors had independent access to the
`data.
`Sample size was based on the results of the European/Cana(cid:173)
`dian MRI Study. 4 It was estimated that 50 evaluable subjects per
`
`Table 1 Baseline demographic and disease characteristics
`
`group would provide 90% power to detect a 60% treatment effect
`between the groups in the total number of GdE lesions at months
`7, 8, and 9 with two-sided a = 0.05.
`
`Results. Follow-up. Subject accrual and follow-up are
`summarized in figure 1. The most common reason for
`screen failure was lack of GdE lesions on screening MRI.
`Three subjects withdrew consent, and one subject did not
`return for the baseline visit. Because of slow accrual, re(cid:173)
`cruitment was discontinued when 90 subjects were en(cid:173)
`rolled, 44 on GA 20 mg and 46 on GA 40 mg. Thirty-nine
`subjects on GA 20 mg and 42 subjects on GA 40 mg had at
`least one MRI scan at month 7, 8, or 9 required for inclu(cid:173)
`sion in the ITT cohort for the primary efficacy endpoint.
`Thirty-eight subjects completed 9 months of double-blind
`treatment on GA 20 mg, and 40 on GA 40 mg. Early
`withdrawal due to adverse events was uncommon and oc(cid:173)
`curred in 1 subject (2.3%) on GA 20 mg vs 4 subjects (8.7%)
`on GA 40 mg (p = 0.36). The subject on 20 mg withdrew
`from the study after experiencing severe dyspnea, speech
`disorder, and panic reaction immediately after injection,
`assessed as related to study medication by the investiga(cid:173)
`tor. The adverse events leading to early termination of 4
`subjects on 40 mg included immediate postinjection reac(cid:173)
`tion (IPIR; n = 2), injection site reaction, (n = 1), and
`increased fatigue (n = 1). There was no difference in time
`to withdrawal due to adverse events (p = 0.95).
`Demographic and baseline disease characteristics. De(cid:173)
`mographic, clinical, and MRI characteristics of the two
`treatment groups were well matched at baseline (table 1).
`The overall study population had active disease with an
`average of 1.5 relapses in the previous year and 3.4 GdE
`lesions at entry.
`MRI outcomes. The results of clinical and MRI efficacy
`analyses are summarized in table 2. Mean total GdE lesion
`number at months 7, 8, and 9 showed a trend favoring GA
`40 mg representing a 38% relative reduction vs GA 20 mg.
`Mean GdE lesions at months 7, 8, and 9 decreased in both
`groups compared with baseline, by 65% in the GA 20 mg
`
`Characteristic
`
`Age, mean (SD), years
`
`Female
`
`White
`
`Years since symptom onset, mean (SD)
`
`Years since diagnosis, mean (SD)
`
`Relapses in previous year, mean (SD)
`
`Actual EDSS score, mean (SD)
`
`Converted EDSS score, mean (SD)
`
`T25FW, mean (SD), seconds
`
`GdE lesion number, mean (SD)
`
`GdE lesion number, median (range)
`
`GdE lesion volume, mean (SD), mL
`
`T2-hyperintense lesion volume, mean (SD), mL
`
`T1-hypointense lesion volume, mean (SD), mL
`
`GA 20 mg
`(n = 44)
`
`37.1 (7.0)
`
`31 (71%)
`
`38 (86%)
`
`7.4 (6.2)
`
`3.2 (3.7)
`
`1.5 (0.8)
`
`2.0 (1.2)
`
`2.0 (1.2)
`
`4.9 (1.3)
`
`3.4 (3.3)
`
`2.0 (1-15)
`
`0.59 (0.686)
`
`16.97 (15.83)
`
`3.65 (6.38)
`
`GA 40 mg
`(n = 46)
`
`37.4 (6.5)
`
`37(80%)
`
`44 (96%)
`
`6.7 (6.4)
`
`3.8 (4.8)
`
`1.5 (0.8)
`
`2.1 (1.0)
`
`2.0 (0.9)
`
`5.1 (1.3)
`
`3.4 (3.1)
`
`2.0 (1-14)
`
`1.17 (3.74)
`
`18.89 (14.71)
`
`4.39 (4.97)
`
`Total
`
`37.2 (6.7)
`
`68 (76%)
`
`82 (91%)
`
`7.1 (6.3)
`
`3.5 (4.3)
`
`1.5 (0.8)
`
`2.0 (1.1)
`
`2.0 (1.0)
`
`5.0 (1.2)
`
`3.4 (3.2)
`
`2.0 (1-15)
`
`0.89 (2.73)
`
`17.96 (15.21)
`
`4.03 (5.67)
`
`GA = glatiramer acetate; EDSS = Expanded Disability Status Scale; T25FW = Timed 25-Foot Walk; GdE = gadolinium-enhancing.
`
`March 20, 2007 NEUROLOGY 68 941
`
`MYLAN PHARMS. INC. EXHIBIT 1056 PAGE 3
`
`

`

`Table 2 Clinical and MRI efficacy results
`
`Endpoint
`
`Primary endpoint
`
`Total number of GdE lesions months 7, 8, 9;
`mean (SD)*
`
`Secondary endpoints
`
`Total number of new GdE lesions months
`8, 9; mean (SD)
`
`Total number of new T2-hyperintense lesions
`months 8, 9; mean (SD)
`
`T2-hyperintense lesion volume change LOV
`vs baseline, mm3
`, adjusted mean (SE)
`
`Total number of confirmed relapses,
`mean (SD)
`
`MSFC change at each visit vs baseline
`
`Prespecified and post hoc exploratory
`endpoints
`
`GdE lesion volume change LOV vs
`baseline, mm3
`, adjusted mean (SE)
`
`T1-hypointense lesion volume change
`LOV vs baseline, mm3
`, adjusted mean (SE)
`
`Number of GdE lesions month 3,
`mean (SD)
`
`Time to first confirmed relapse
`(20th percentile), days
`
`EDSS change at each visit vs baseline
`
`Respond erst
`
`Responders:j:
`
`GA 20 mg
`(n = 44)
`
`GA 40 mg
`(n = 46)
`
`Effect size (95% CI)
`
`p Value
`
`3.62 (4.06)
`
`2.26 (4.06)
`
`Rate ratio 0.62 (0.36, 1.08)
`
`0.0898
`
`1.41 (1.86)
`
`1.00 (1.91)
`
`Rate ratio 0. 73 (0.39, 1.35)
`
`1.38 (1.76)
`
`1.00 (2.00)
`
`Rate ratio 0. 70 (0.38, 1.30)
`
`800 (1,144)
`
`1,516 (1,095)
`
`Difference 715 ( -3,678, 2,248)
`
`0.52 (0.59)
`
`0.30 (0.59)
`
`Rate ratio 0.59 (0.31, 1.16)
`
`0.311
`
`0.256
`
`0.631
`
`0.121
`
`No change or between-group
`difference
`
`-684 (50.81)
`
`-801 (49.55)
`
`Difference 117 ( -16, 251)
`
`0.0841
`
`122.29 (115.71)
`
`32.23 (109.72)
`
`Difference 90.16 ( -208.10, 388.42)
`
`0.548
`
`2.61 (4.22)
`
`1.33 (1.58)
`
`Rate ratio 0.48 (0.29, 0.82)
`
`0.0051
`
`0.0183
`
`0.0367
`
`Relapse-free subjects
`
`23/44 (52.3%)
`
`35/46 (76.1%)
`
`Risk ratio 0.50 (0.27, 0.91)
`
`NNT = 1.9
`
`NNT = 1.3
`
`80
`
`213
`
`No change or between-group
`difference
`
`15/39 (38.5%)
`
`29/42 (69.0%)
`
`Odds ratio 3.51 (1.39, 8.88)
`
`0.0078
`
`NNT = 2.6
`
`NNT = 1.5
`
`5/37 (13.5%)
`
`13/40 (32.5%)
`
`Odds ratio 3.12 (1.00, 11.13)
`
`0.0049
`
`NNT = 7.4
`
`NNT = 3.1
`
`39 of 44 subjects on 20 mg and 42 of 46 subjects on 40 mg had at least one MRI scan at month 7, 8, or 9.
`t Relapse free with no gadolinium-enhancing (GdE) lesions at months 7, 8, and 9 or a reduction in the mean number reduced by at
`least 50% vs baseline.
`:j: Relapse free, no Expanded Disability Status Scale (EDSS) progression, no GdE lesions at months 7, 8, and 9; and no new T2-
`hyperintense lesions at the last assessment vs baseline.
`
`GA = glatiramer acetate; LOV = last observed value; MSFC = multiple sclerosis functional composite; NNT = number needed to treat.
`
`group and 75% in the GA 40 mg group (p < 0.0001 for
`both; figure 2). The advantage of 40 mg over 20 mg was
`apparent as early as month 3 (figure 2). Trends favoring 40
`mg were seen in change from baseline to last observed
`value of GdE lesion volume, total number of new GdE
`lesions at months 8 and 9, and new T2-hyperintense le(cid:173)
`sions at months 8 and 9. There were no differences in
`change from baseline to last observation in total T2-
`hyperintense or Tl-hypointense lesion volumes.
`Clinical outcomes. On-study relapse rate decreased in
`both groups compared with the previous year. Mean on(cid:173)
`study relapse rate showed a trend favoring GA 40 mg, with
`a greater proportion of relapse-free subjects and delay in
`the time to first confirmed relapse. EDSS and T25FW did
`
`942 NEUROLOGY 68 March 20, 2007
`
`not change in either treatment group, and there were no
`between-group differences at any time point. In a post hoc
`analysis, two definitions of treatment responders were
`used, both showing benefit favoring GA 40 mg.
`Safety and tolerability. Safety profiles of the two GA
`doses were similar. There were no deaths or significant
`effects on vital signs. There were two serious adverse
`events. A subject treated with GA 40 mg was hospitalized
`after an IPIR and subsequently discontinued from the
`study. A subject treated with GA 20 mg was involved in a
`motor vehicle accident, classified as unrelated to study
`drug.
`Injection site reactions were the most frequent adverse
`event for both doses, occurring in 38 (86.4%) subjects in
`
`MYLAN PHARMS. INC. EXHIBIT 1056 PAGE 4
`
`

`

`····+···· 20mg
`
`--40mg
`
`······-t················ .. ....
`
`0
`
`23 456 78 9
`Months
`
`(/)
`+I
`c
`
`w 5.0
`4.5
`4.0
`"' CD
`.s
`3.5
`w 3.0
`.0
`E 2.5
`:J z
`2.0
`c
`1.5
`0
`'iii
`CD
`1.0
`_j
`UJ
`0.5
`"0
`(9
`0
`
`!::>~<:>;
`"'"'
`c:Jl'
`Figure 2. Gadolinium-enhancing (GdE) lesion number at
`each visit.
`
`the 20-mg group and 39 (84.8%) in the 40-mg group. Injec(cid:173)
`tion site manifestations reported with at least 5% higher
`incidence for GA 40 mg included burning, mass, pain, and
`urticaria (table 3). Skin necrosis and lipoatrophy were not
`observed. Thirty-nine IPIRs occurred in 10 (22.7%) sub(cid:173)
`jects on GA 20 mg vs 52 IPIRs in 15 (32.6%) subjects on
`GA 40 mg. The incidence in the 20-mg group was consis(cid:173)
`tent with previous large studies of GA. 1
`3 IPIRs were cate(cid:173)
`•
`gorized most often as moderate severity in the 40-mg
`group and mild in the 20-mg group. All IPIRs resolved
`without sequelae, although one in the 40-mg group led to
`hospitalization and subsequent discontinuation from the
`study. The greatest difference among IPIR component
`symptoms (flushing, palpitations, tachycardia, dyspnea,
`and chest pain) was in palpitations. Affect lability was
`reported by 6 subjects (13%) treated with GA 40 mg and by
`none in the GA 20 mg group. All cases were assessed as
`not related to study drug by the investigator, and in no
`case was study drug discontinued.
`Postbaseline shifts to abnormal laboratory values were
`seen for white blood cell count, platelets, eosinophils, glu-
`
`Table 3 Number and percentage of subjects experiencing adverse
`events with frequency differing by 5% or greater in the 40-mg
`group compared with the 20-mg group
`
`Adverse event
`
`Injection site burning
`
`Injection site mass
`
`Injection site pain
`
`Urticaria
`
`Any symptom of IPIR
`
`Palpitations
`
`Flushing
`
`Affect lability
`
`Muscle cramp
`
`Pharyngitis
`
`Headache
`
`Hypoesthesia
`
`Paresthesia
`
`GA 20 mg
`(n = 44)
`
`6 (13.6%)
`
`9 (20.5%)
`
`9 (20.5%)
`
`1 (2.3%)
`
`10 (22.7%)
`
`1 (2.3%)
`
`6 (13.6%)
`
`0
`
`0
`
`0
`
`4 (9.1%)
`
`4 (9.1%)
`
`6 (13.6%)
`
`GA 40 mg
`(n = 46)
`
`13 (28.3%)
`
`16 (34.8%)
`
`14 (30.4%)
`
`5 (10.9%)
`
`15 (32.6%)
`
`5 (10.9%)
`
`9 (19.6%)
`
`6 (13%)
`
`3 (6.5%)
`
`3 (6.5%)
`
`7 (15.2%)
`
`7 (15.2%)
`
`1 (2.2%)
`
`GA = glatiramer acetate; IPIR = immediate postinjection
`reaction.
`
`cose, cholesterol, and liver enzymes. In general, these ab(cid:173)
`normalities were transient, mild, and uncommon and
`occurred with equal incidence in the two groups. Most of
`the potentially clinically significant laboratory abnormali(cid:173)
`ties seemed to be related to concomitant medical condi(cid:173)
`tions. An unexpected finding of isolated occurrences of
`hypocalcemia in approximately 18% of subjects in both
`groups was without clear-cut explanation. No EKG abnor(cid:173)
`mality was attributed to GA.
`
`Discussion. The safety and efficacy of GA at the
`currently approved 20-mg daily dose are supported
`by three pivotal trials, 1
`4 a meta-analysis of those
`-
`studies,S a long-term follow-up study,6 and postmar(cid:173)
`keting experience. This 9-month multicenter, ran(cid:173)
`domized,
`double-blind,
`parallel-group
`trial
`represents the first dose-comparison study of GA.
`There was a trend favoring the 40-mg dose on the
`primary endpoint, total number of GdE lesions at
`months 7, 8, and 9, which was supported by signifi(cid:173)
`cant results or trends favoring 40 mg on other
`secondary and exploratory MRI endpoints, relapse(cid:173)
`related endpoints, and responder analyses.
`There have been three previous trials of the MRI
`effects of GA. In a small, single-arm crossover
`study, 15 Mancardi et al. demonstrated a 57% reduc(cid:173)
`tion in the frequency of new GdE lesions during 20
`mg GA treatment compared with the pretreatment
`baseline period. In an ancillary study of 27 subjects
`enrolled at one site in the US pivotal trial in
`RRMS, 16 GA treatment produced significant reduc(cid:173)
`tions in GdE lesions and brain volume loss. The most
`definitive study was the European/Canadian MRI
`trial, 4 which demonstrated significant reduction in
`cumulative GdE lesions on monthly MRI scans over
`9 months favoring GA 20 mg over placebo. Consis(cid:173)
`tent differences favoring active treatment were seen
`on other GdE-related
`endpoints,
`new T2-
`hyperintense lesions, and relapse rate (33% reduc(cid:173)
`tion). Treatment with GA also
`reduced
`the
`proportion of new lesions that evolved into chronic
`Tl-hypointense lesions ("black holes"), 17 which are
`thought to indicate more severe tissue damage. 18,19
`The design of the current trial was modeled after
`the European/Canadian MRI trial. In general, sub(cid:173)
`jects treated with GA 20 mg in the current study
`fared somewhat better than the GA 20 mg group in
`the European/Canadian MRI trial. For example, the
`mean number of GdE lesions at termination was re(cid:173)
`duced by approximately 65% compared with 48% in
`the previous study. This variability probably is due
`to differences in study populations. Subjects enrolled
`in the European/Canadian MRI trial had more active
`disease as indicated by a mean of 4.2 GdE lesions on
`the baseline MRI vs 3.4. Other factors also may have
`played a role, including different "regression to the
`mean" or unknown confounders.
`In the European!Canadian MRI trial, the mean
`and median cumulative numbers of GdE lesions in
`the two treatment groups seemed to diverge between
`months 3 and 6. 4 This result has been interpreted as
`March 20, 2007 NEUROLOGY 68 943
`
`MYLAN PHARMS. INC. EXHIBIT 1056 PAGE 5
`
`

`

`suggesting a 3- to 6-month lag in the onset of action
`of GA 20 mg on MRI -detected disease activity. In the
`present study, an advantage of the 40-mg dose on
`suppressing GdE lesions was already evident at
`month 3 and, in fact, was more prominent compared
`with months 7, 8, and 9. Similarly, results based on
`the first relapse, e.g., time to first relapse and pro(cid:173)
`portion of relapse-free subjects, were more robust
`than measures based on the 9-month study period,
`e.g., relapse rate. These results suggest that the on(cid:173)
`set of action of the 40-mg dose is more rapid com(cid:173)
`pared with 20 mg. A larger, longer study will be
`necessary to confirm the sustainability of the efficacy
`advantage of the higher dose.
`GA at the currently approved 20-mg dose has been
`safe and well tolerated in previous trials and post(cid:173)
`marketing experience. The overall safety and side
`effect profile of the 40-mg dose in this trial was sim(cid:173)
`ilar, although it was associated with a greater inci(cid:173)
`dence of certain adverse effects. For example,
`although the overall incidence of injection site reac(cid:173)
`tions was similar, some aspects were more common
`with the higher dose, and the injections seemed to be
`somewhat more painful. IPIRs also were somewhat
`more common and severe. Qualitatively, both side
`effects were similar
`to what has been seen
`previously.
`The other adverse event of note was affect lability
`reported in 13% of subjects treated with 40 mg but
`none on 20 mg. Mood disorders are common in MS,
`and a relationship between mood disorders and GA
`20 mg has not been seen in previous trials or post(cid:173)
`marketing surveillance. In the US pivotal trial in
`RRMS, affect lability occurred with a higher inci(cid:173)
`dence in the placebo group. All of the cases in the
`current study were assessed as unrelated to study
`drug by the site investigator. In no case was study
`drug discontinued because of this adverse effect.
`Thus, a causal relationship to the 40-mg dose is un(cid:173)
`likely. Nevertheless, this issue needs to be assessed
`further in other studies of the higher dose.
`
`References
`1. Bornstein MB, Miller A, Slagle S, et al. A pilot trial of Cop 1 in
`exacerbating-remitting multiple sclerosis. N Eng! J Med 1987;317:408-
`414.
`
`2. Johnson KP, Brooks BR, Cohen JA, et al. Copolymer 1 reduces relapse
`rate and improves disability in relapsing-remitting multiple sclerosis:
`results of a phase III multicenter, double-blind, placebo-controlled trial.
`Neurology 1995;45:1268-1276.
`3 .• Tohnson KP, Brooks BR, Cohen JA, et al. Extended use of glatiramer
`acetate (Copaxone) is well tolerated and n1aintains its clinical efiect on
`multiple sclerosis relapse rate and degree of disability. Neurology 1998;
`50:701-708.
`4. Comi G, Filippi M, Wolinsky .TS, the European/Canadian Gl