Multiple Sclerosis and Related Disorders 4 (2015) 370–376
`
`Contents lists available at ScienceDirect
`
`Multiple Sclerosis and Related Disorders
`
`journal homepage: www.elsevier.com/locate/msard
`
`GLACIER: An open-label, randomized, multicenter study to assess the
`safety and tolerability of glatiramer acetate 40 mg three-times weekly
`versus 20 mg daily in patients with relapsing-remitting multiple
`sclerosis
`Jerry S. Wolinsky a,n, T. Erik Borresen b, Dennis W Dietrich c, Daniel Wynn d, Yulia Sidi e,
`Joshua R. Steinerman f, Volker Knappertz f,g, Scott Kolodny h, on behalf of the GLACIER
`Study Group1
`a University of Texas Health Science Center at Houston, Houston, TX, USA
`b Mecklenburg Neurological Associates, Charlotte, NC, USA
`c Advanced Neurological Specialists, Great Falls, MT, USA
`d Consultants in Neurology, Northbrook, IL, USA
`e Teva Pharmaceuticals, Netanya, Israel
`f Global Clinical Development, Teva Pharmaceuticals, Frazer, PA, USA
`g Heinrich‐Heine Universität Düsseldorf, Düsseldorf, Germany
`h Teva Pharmaceuticals, Cleveland, OH, USA
`
`a r t i c l e i n f o
`
`a b s t r a c t
`
`Article history:
`Received 19 February 2015
`Received in revised form
`5 June 2015
`Accepted 11 June 2015
`
`Keywords:
`Glatiramer acetate
`Relapsing-remitting multiple sclerosis
`Dose regimen
`Safety
`Convenience
`Patient adherence
`
`Background: The efficacy and safety of glatiramer acetate (GA) 20 mg/mL once-daily subcutaneous injections
`(GA20) in relapsing-remitting multiple sclerosis (RRMS) is well-established. However, injection-related ad-
`verse events (IRAEs) may impede treatment adherence and tolerability. GA 40 mg/mL three-times weekly
`(GA40) also has a favorable efficacy and safety profile.
`Objective: To evaluate the safety, tolerability, and patient experience when converting from GA20 to GA40.
`Methods/trial design: GLACIER was an open-label, randomized, parallel-group trial conducted at 31 sites in the
`US between June 2013 and December 2013. Stable RRMS patients on GA20 were randomized in a 1:1 ratio to
`continue with GA20 or convert to GA40. The adjusted mean annualized rate of IRAEs was the primary endpoint
`for this study. Additionally, the severity of IRAEs, rate of injection-site reactions (ISRs), and patient-reported MS
`impact and treatment satisfaction were compared for the two treatment groups over the 4-month core study.
`Results: A total of 209 patients were randomized to convert to GA40 (n¼108) or continue with GA20
`(n¼101). The adjusted mean annualized rate of IRAEs was reduced by 50% with GA40 (35.3 events per year;
`n¼108) versus GA20 (70.4 events per year; n¼101) (risk ratio (RR)¼0.50; 95% confidence interval [CI]¼
`0.34–0.74; p¼0.0006). There was a 60% reduction in the rate of moderate/severe events (GA40 (n¼108):
`0.9 events per year versus GA20 (n¼101): 2.2 events per year; RR¼0.40; p¼0.0021). Perception of treatment
`convenience improved for GA40-treated patients soon after converting and was sustained.
`Conclusions: The GLACIER study demonstrates a favorable IRAE and convenience profile of GA40 for RRMS
`patients.
`Trial registration: NCT01874145 available at clinicaltrial.gov.
`& 2015 The Authors. Published by Elsevier B.V. This is an open access article under the CC BY-NC-ND license
`(http://creativecommons.org/licenses/by-nc-nd/4.0/).
`
`1.
`
`Introduction
`
`Disease-modifying therapies (DMTs) that reduce the frequency
`of relapses, reduce accumulation of disability, and control symp-
`toms have improved the care of patients with relapsing-remitting
`
`n Corresponding author. Fax: þ1 713 500-7040.
`E-mail address: Jerry.S.Wolinsky@uth.tmc.edu (J.S. Wolinsky).
`1 See Supplementary material for a listing of investigators and sites.
`
`multiple sclerosis (RRMS) (Damal et al., 2013; Compston and Coles,
`2002). Glatiramer acetate (GA), a first-line therapy approved for
`the treatment of RRMS (TEVA Neuroscience, Inc. 2014), has a well-
`characterized long-term safety profile (TEVA Neuroscience, Inc.
`2014; Boster et al., 2011), with more than 2 million patient-years of
`overall exposure to GA 20 mg/mL administered once daily by
`subcutaneous injection (GA20) (data on file), and reduces the
`frequency of relapses and magnetic resonance imaging (MRI)
`disease activity (Johnson et al., 1995; Martinelli Boneschi et al.,
`
`http://dx.doi.org/10.1016/j.msard.2015.06.005
`2211-0348/& 2015 The Authors. Published by Elsevier B.V. This is an open access article under the CC BY-NC-ND license (http://creativecommons.org/licenses/by-nc-nd/4.0/).
`
`Mylan Pharms. Inc. Exhibit 1038 Page 1
`
`

`

`J.S. Wolinsky et al. / Multiple Sclerosis and Related Disorders 4 (2015) 370–376
`
`371
`
`2003; Comi et al., 2001; Bornstein et al., 1987).
`Several first-line treatment regimens for RRMS require long-
`term, frequent injection of the drug, making adherence a challenge
`for many patients despite satisfactory efficacy (Verdun di Can-
`togno et al., 2011; Devonshire et al., 2011). Factors contributing to
`non-adherence and reduced tolerability in MS treatment include
`problems with injecting, perceived lack of efficacy, and perhaps
`most importantly, the high incidence of injection-related adverse
`events (IRAEs) (Devonshire et al., 2011; Treadaway et al., 2009;
`Costello et al., 2008). IRAEs associated with injectable DMTs and
`GA include systemic immediate post-injection reactions (IPIRs)
`such as flushing and anxiety and, more often, local injection-site
`reactions (ISRs), such as pain and redness (TEVA Neuroscience, Inc.
`2014). More severe ISRs, such as lipoatrophy and skin necrosis,
`occur less frequently (Costello et al., 2008). Modified treatment
`regimens – alternative dosages and low-frequency administration
`schedules – have the potential to reduce the rate and severity of
`IRAEs. Therefore, modifying the treatment regimen of drugs with
`proven, long-term efficacy can result in better adherence and
`tolerability while maintaining efficacy and improving treatment
`convenience and patient experience (Remington et al., 2013; Tan
`et al., 2011). The value of such alternative regimens is enhanced
`given the importance of treatment adherence in ensuring optimal
`clinical outcomes (Tan et al., 2011; Al-Sabbagh et al., 2008).
`In 2014, based largely on the results of the Glatiramer Acetate
`Low-frequency Administration (GALA) study, GA 40 mg/mL ad-
`ministered three-times weekly by subcutaneous injection (GA40)
`was approved for the treatment of RRMS by regulatory authorities
`in an increasing number of countries worldwide. The low-fre-
`quency GA40 was shown to have favorable efficacy and safety
`profiles (Khan et al., 2013).
`Previous GA40 trials only enrolled patients who were naïve to
`GA treatment and did not examine those converting from GA20 to
`GA40. The GLACIER (GLatiramer Acetate low frequenCy safety and
`patIent ExpeRience) study was performed to assess the safety and
`tolerability of GA40 compared with GA20 in clinically stable pa-
`tients who had been treated continuously with GA20 for a mini-
`mum of 6 months before screening. This study provides insight
`into whether GA40 provides improved safety, tolerability, and
`patient experience compared with the established GA20 regimen.
`
`2. Materials and methods
`
`2.1. Study design and patients
`
`The GLACIER study was an open-label, randomized, parallel-group
`study conducted at 31 sites in the United States between June 2013
`and December 2013. Investigators were neurologists, and sites con-
`sisted of individual and group neurology practices, neurology and MS
`research centers, and independent clinic trial facilities. All institutional
`review boards or ethics committees of the participating centers ap-
`proved the protocol, and all patients gave written informed consent
`before any study-related procedures were performed.
`Key eligibility criteria required patients to be least 18 years of
`age, with a confirmed and documented RRMS diagnosis (according
`to the revised McDonald criteria (Polman et al., 2011)) and an
`Expanded Disability Status Scale (EDSS) score of r5.5 at screening
`and baseline visits. All patients were required to be on continuous
`GA20 treatment for Z6 months before screening and to be neu-
`rologically stable and relapse-free for Z60 days before
`randomization.
`Patients with progressive forms of MS, or those with neuro-
`myelitis optica, were excluded. Other exclusion criteria included
`treatment with experimental or
`investigational drugs; con-
`comitant use of other MS disease-modifying drugs; chronic (430
`
`days) systemic corticosteroid treatment within 6 months of
`screening; and prior use of mitoxantrone, cladribine, alemtuzu-
`mab, rituximab, or natalizumab.
`At the baseline visit, eligible patients were randomized in a 1:1
`ratio to either continue with GA20 or convert to GA40. The com-
`puterized randomization sequence was generated and maintained
`by the Clinical Supply Chain department at Teva Pharmaceuticals
`(Netanya,
`Israel), and randomization was conducted centrally
`using the Interactive Response Technology system. Patients were
`randomized according to the randomization scheme of con-
`strained blocks by site, and treatment group assignment was not
`biased by patient or trial center preferences. Investigators and
`participants were not blinded or masked to the open-label treat-
`ment assignment. Patients were treated with either a single-use,
`pre-filled syringe containing GA20 or GA40 (Teva Pharmaceutical
`Industries) in a 1-mL suspension of 40 mg of mannitol USP/Ph.Eur
`dissolved in water. Five scheduled site visits occurred during the
`core phase at months –1 (screening), 0 (baseline), 1, 2, and 4
`(termination). Eligible patients from both treatment arms who
`completed the core phase were offered the opportunity to parti-
`cipate in an extension phase, during which they would receive
`GA40 treatment.
`
`2.2. Procedures
`
`The primary endpoint was the rate of IRAEs in each treatment
`arm. Secondary endpoints included the rate of ISRs, patient-re-
`ported impact of MS on physical and psychological well-being
`using the Multiple Sclerosis Impact Scale-29 (MSIS-29) ques-
`tionnaire, and patient perceptions of convenience and overall sa-
`tisfaction using subscales of the Treatment Satisfaction Ques-
`tionnaire for Medication-9 (TSQM-9). Additional endpoints in-
`cluded baseline patient expectations for efficacy, safety, and con-
`venience of GA40 compared with GA20.
`Assessments of IRAEs were performed throughout the study
`based on the patient's diary card recordings of occurrence and
`severity of IRAEs. IRAEs included all local ISRs or symptoms or
`events related to IPIRs, such as flushing, chest pain, palpitations,
`anxiety, dyspnea, throat constriction, or urticaria. Severity was
`defined by the patient as mild if the IRAE is ‘easily tolerated,’
`moderate if the IRAE ‘interferes with normal daily activity’, or
`severe if the IRAE ‘prevents normal daily activity’.
`Study drug compliance was evaluated during each visit after the
`initial dispensation of the study drug, and study drug accountability
`records were completed. Compliance was calculated as a percentage
`by dividing the number of used syringes by the number of total
`syringes expected to be used, multiplied by 100. The incidence of
`patients in each arm of the study with Z75% overall compliance to
`the study drug was an exploratory trial outcome.
`The MSIS-29 questionnaire and TSQM-9 were performed at
`baseline and Months 1, 2, and 4. The validated, 29-item MSIS-29
`questionnaire was used to assess patient-reported impact of MS on
`physical well-being and psychological well-being. Responses were
`scored using a five-point Likert scale range, with higher aggregate
`scores corresponding to greater impact on well-being. The vali-
`dated TSQM-9 was used to assess patient-reported perceptions of
`convenience (items 4–6) and overall satisfaction (items 1–3), with
`higher scores representing more positive perceptions.
`Patients' expectations of convenience were assessed at baseline
`using a study-specific questionnaire, in which patients reported
`whether they expected GA40 to be less, equally, or more con-
`venient compared with GA20. Similarly, patients reported whether
`they expected GA40 to be less, equally, or more safe, as well as
`effective, compared with GA20.
`Safety assessments included adverse events (AEs), vital signs,
`electrocardiographic (ECG) measurements, and standard clinical
`
`Mylan Pharms. Inc. Exhibit 1038 Page 2
`
`

`

`372
`
`J.S. Wolinsky et al. / Multiple Sclerosis and Related Disorders 4 (2015) 370–376
`
`laboratory tests, including complete blood count and compre-
`hensive metabolic panel.
`
`2.3. Statistical analyzes
`
`A sample size of 200 patients (100 patients per arm) was
`considered necessary to provide 80% power to detect a statistically
`significant difference in the IRAE rates between treatment arms.
`This calculation utilized estimated event rates from previous stu-
`dies of each treatment regimen and accounted for an expected
`relative risk of 0.48 (52% reduction in the risk for IRAEs) in the
`GA40-treated population compared with GA20 and a dropout rate
`of 15%. The overall significance level for this study was 5% using a
`two-tailed test, and to protect the study from Type I error inflation,
`the secondary endpoints were only analyzed if a statistically sig-
`nificant treatment effect was detected in the primary analysis.
`The principal analysis for the primary endpoint of rate of IRAEs
`during the core phase was performed on the safety analysis co-
`hort, defined as all randomized patients who received at least one
`dose of the study drug. For each patient, multiple IRAEs starting on
`the same day were only counted as one IRAE event. The GA40
`group was compared with the GA20 group using a baseline-ad-
`justed, quasi-likelihood (over-dispersed) Poisson regression with
`the natural log of treatment duration (in years) as an offset vari-
`able. In addition to treatment group, the following covariates were
`included in the regression model: baseline EDSS score, age, gender,
`and number of relapses in the 2 years before screening. A similar
`post hoc analysis assessed risk ratio (RR) for annualized rate of
`moderate or severe IRAEs.
`To keep the overall Type I error of 5% from inflating, a gate-
`keeping approach was utilized for secondary endpoints, with a
`pre-specified hierarchical sequence: 1, ISR rate; 2, MSIS-29 phy-
`sical score change from baseline to Month 4; 3, MSIS-29 psycho-
`logical score change from baseline to Month 4; 4, TSQM-9 con-
`venience score change from baseline to Month 4; and 5, TSQM-9
`satisfaction score change from baseline to Month 4.
`The secondary endpoint of rate of ISRs in each treatment group
`was assessed using the same cohort and statistical analysis method
`as the IRAE rate. Secondary analyzes for the other secondary end-
`points were performed on the full analysis cohort, defined as all
`randomized patients who received at least one dose of the study
`drug and had at least one post-baseline assessment of either the
`MSIS-29 or TSQM-9. Change from baseline to Month 4 in physical
`well-being and psychological well-being using the MSIS-29, and
`patient perceptions of convenience and overall satisfaction using the
`TSQM-9, were analyzed using mixed model repeated measures
`analysis of covariance. In addition to treatment group, the following
`covariates were used in this model: baseline questionnaire score,
`categorical month, and treatment by month interaction. Additionally,
`month was specified as a repeated effect, and an unstructured cor-
`relation matrix was used to model intra-subject correlation.
`Analysis of patient-reported expectations for GA40 at baseline
`was performed on the full analysis cohort, and patient counts and
`percentages were provided.
`
`3. Results
`
`3.1. Patients
`
`Of the 218 patients screened for entry into the GLACIER study,
`209 were randomized to the study treatment (GA20, n¼101;
`GA40, n¼108) (Fig. 1). Baseline demographics showed no dis-
`cernable differences between the two treatment arms (Table 1).
`The proportions of patients who prematurely discontinued treat-
`ment were similar for GA20 (3.0%, n¼3) and for GA40 (6.5%, n¼7)
`
`groups, and one patient in the GA40 treatment arm prematurely
`discontinued the study because of injection-site necrosis (Fig. 1).
`Overall compliance with study medication use was good, with
`100% of GA20 patients and 99% GA40 patients in the ITT cohort
`taking 475% of all expected doses. The safety analysis cohort
`consisted of all 209 patients, and the full analysis cohort consisted
`of 208 patients (GA20, n¼100; GA40, n¼108). Of the 98 patients
`who completed GA20 treatment, 97 (99%) elected to participate in
`the extension phase; all 101 of the patients who completed GA40
`treatment (100%) elected to continue GA40 treatment during the
`extension phase.
`
`3.2. Rate of IRAEs
`
`Patients receiving GA40 demonstrated a 50% lower annualized
`IRAE rate than those receiving GA20 (adjusted mean annualized
`event rate, 70.4 vs 35.3 events per year; RR¼0.50; 95% confidence
`interval [CI]¼0.34–0.74; p¼0.0006) (Fig. 2). The total number of
`individually reported IRAEs on GA40 was 2670; on GA20, there
`were 5770 IRAEs. Of these, moderate or severe reactions were less
`frequent in GA40-treated patients (9.1%) than in GA20-treated
`patients (18.8%) (Table 2). Post hoc analyzes showed that patients
`receiving GA40 demonstrated a 60% reduction in the annualized
`event rate of moderate/severe IRAEs compared with GA20 (ad-
`justed mean annualized event rate, 0.9 vs 2.2 events per year;
`RR¼0.40; 95% CI¼0.23–0.72; p¼0.0021) (Fig. 3). The proportion
`of patients experiencing at least one IRAE was similar between
`treatment arms (56.4% for GA20 versus 58.3% for GA40) (Table 2).
`
`3.3. Rate of ISRs
`
`Most IRAEs were ISRs (Table 2), and consequently, the ISR and
`IRAE rate reductions for GA40 versus GA20 were almost identical.
`Patients receiving GA40 demonstrated a 50% reduction in the an-
`nualized ISR rate compared with GA20 (mean annualized event
`rate, 70.4 vs 35.2 events per year; RR¼0.50; 95% CI¼0.34–0.74;
`p¼0.0006) (Fig. 4). The total number of individually reported ISRs
`on GA40 was 2610 compared to 5537 events with GA20. Of these,
`moderate or severe ISRs were less frequent in GA40-treated pa-
`tients (9.2%) than in GA20-treated patients (15.4%) (Table 2).
`The reduced ISR frequency and event rates on GA40 were ob-
`served in all common ISR types, as categorized by MedDRA 16.0
`preferred terms (Table 3). In particular, common ISR events such as
`
`Fig. 1. Patient disposition. GA20, glatiramer acetate 20 mg/mL daily; GA40, glatir-
`amer acetate 40 mg/mL three-times weekly.
`
`Mylan Pharms. Inc. Exhibit 1038 Page 3
`
`

`

`J.S. Wolinsky et al. / Multiple Sclerosis and Related Disorders 4 (2015) 370–376
`
`373
`
`Table 1
`Baseline patient and disease characteristics.
`
`Age, years, mean (7SD)
`Female, n (%)
`Race
`Caucasian
`Black/African American
`Asian
`Native American/Alaskan Native
`Body mass index, mean (7SD)
`EDSS, mean (7SD)
`Years from onset of first MS symptoms,
`mean (7SD)
`Years from MS diagnosis, mean (7SD)
`Exacerbations over 1 year prior to study
`initiation, mean (7SD)
`Exacerbations over 2 years prior to study
`initiation, mean (7SD)
`Years of GA20 exposure prior to study
`initiation, mean (7SD)
`
`GA20 (n¼101) GA40 (n¼108)
`
`50.4 (9.3)
`83 (82.2)
`
`96 (95.0)
`5 (5.0)
`0 (0.0)
`0 (0.0)
`29.3 (6.4)
`2.4 (1.4)
`16.2 (11.0)
`
`50.9 (11.0)
`89 (82.4)
`
`100 (92.6)
`5 (4.6)
`3 (2.8)
`0 (0.0)
`27.9 (6.2)
`2.5 (1.4)
`15.7 (11.1)
`
`12.1 (10.0)
`0.2 (0.4)
`
`10.8 (8.6)
`0.2 (0.5)
`
`0.4 (0.7)
`
`0.4 (0.6)
`
`6.6 (4.3)
`
`6.9 (5.2)
`
`EDSS: Expanded Disability Status Scale, GA20: glatiramer acetate 20 mg/mL daily,
`GA40: glatiramer acetate 20 mg/mL three-times weekly, MS: multiple sclerosis, SD:
`standard deviation.
`
`Fig. 2. Annualized rate of IRAEs. *Event rate per year equals the total number of
`events/exposure to study drug (in years). Adjusted mean and other statistics were
`derived from a baseline-adjusted, quasi-likelihood (over-dispersed) Poisson re-
`gression model with natural log of treatment duration (in years as an offset vari-
`able, adjusted for baseline Expanded Disability Status Scale score, treatment group,
`age, sex, and number of relapses in the 2 years prior to screening. CI: confidence
`interval, GA20: glatiramer acetate 20 mg/mL daily, GA40: glatiramer acetate
`40 mg/mL three-times weekly, IRAE: injection-related adverse event, RR: risk ratio.
`
`Table 2
`IRAE and ISR frequency and severity.
`
`GA20 (n¼101) GA40 (n¼108)
`
`Total number of IRAEsa
`Number of moderate or severe IRAEs (% of
`total)
`Patients with Z1 IRAE, n (%)
`Total number of ISRsa
`Number of moderate or severe ISRs (% of
`total)
`Patients with Z1 ISR, n (%)
`
`5770
`1083 (18.8)
`
`57 (56.4)
`5537
`854 (15.4)
`
`2670
`244 (9.1)
`
`63 (58.3)
`2610
`240 (9.2)
`
`57 (56.4)
`
`61 (56.5)
`
`GA20: glatiramer acetate 20 mg/mL daily, GA40: glatiramer acetate 40 mg/mL
`three-times weekly,
`IRAE:
`injection-related adverse event,
`ISR:
`injection-site
`reaction
`a The number of IRAEs and ISRs represent individually reported events.
`
`Fig. 3. Annualized rate of moderate or severe IRAEs. *Event rate per year equals the
`total number of events/exposure to study drug (in years). Adjusted mean and other
`statistics were derived from a baseline-adjusted, quasi-likelihood (over-dispersed)
`Poisson regression model with natural log of treatment duration (in years as an
`offset variable, adjusted for baseline Expanded Disability Status Scale score, treat-
`ment group, age, sex, and number of relapses in the 2 years prior to screening. CI:
`confidence interval, GA20: glatiramer acetate 20 mg/mL daily, GA40: glatiramer
`acetate 40 mg/mL three-times weekly, IRAE: injection-related adverse event, RR:
`risk ratio.
`
`injection-site pain, erythema, mass, swelling, and pruritus had no-
`tably lower rates of occurrence among GA40-treated patients. The
`rates of injection-site atrophy and necrosis were minimal (r0.1
`events per year), with no observable difference between treatment
`groups. Furthermore, the annualized IPIR rate with GA40 (1.9 events
`per year) was lower than that of GA20 (3.9 events per year).
`
`3.4. Patient-reported impact on physical and psychological well-
`being using the MSIS-29
`
`Although the MSIS-29 physical score change from baseline to
`Month 4 showed a trend toward improvement in the GA40 treat-
`ment group versus the GA20 group, this outcome did not reach
`
`Fig. 4. Annualized rate of ISRs. *Event rate per year equals the total number of
`events/exposure to study drug (in years). Adjusted mean and other statistics were
`derived from a baseline-adjusted, quasi-likelihood (over-dispersed) Poisson re-
`gression model with natural log of treatment duration (in years as an offset vari-
`able, adjusted for baseline Expanded Disability Status Scale score, treatment group,
`age, sex, and number of relapses in the 2 years prior to screening. CI: confidence
`interval, GA20: glatiramer acetate 20 mg/mL daily, GA40: glatiramer acetate
`40 mg/mL three times weekly, ISR: injection-site reaction, RR: risk ratio.
`
`Mylan Pharms. Inc. Exhibit 1038 Page 4
`
`

`

`374
`
`J.S. Wolinsky et al. / Multiple Sclerosis and Related Disorders 4 (2015) 370–376
`
`Table 3
`ISR frequency and annualized rates.
`
`Number of ISRs (annualized event rate)
`
`GA20 (n¼101)
`
`GA40 (n¼108)
`
`Injection-site atrophy
`Injection-site bruising
`Injection-site discoloration
`Injection-site discomfort
`Injection-site dryness
`Injection site erythema
`Injection-site hematoma
`Injection-site hemorrhage
`Injection-site induration
`Injection site inflammation
`Injection-site irritation
`Injection-site mass
`Injection-site necrosis
`Injection-site pain
`Injection site pruritus
`Injection-site rash
`Injection-site reaction
`Injection-site swelling
`Injection-site urticaria
`Injection-site vesicles
`Injection-site warmth
`
`0 (0.0)
`166 (5.4)
`4 (0.1)
`4 (0.1)
`0 (0.0)
`1331 (43.5)
`1 (0.0)
`43 (1.4)
`2 (0.1)
`1 (0.0)
`0 (0.0)
`669 (21.9)
`0 (0.0)
`1692 (55.3)
`398 (13.0)
`13 (0.4)
`0 (0.0)
`594 (19.4)
`377 (12.3)
`14 (0.5)
`228 (7.5)
`
`1 (0.0)
`8 (0.2)
`9 (0.3)
`0 (0.0)
`48 (1.5)
`692 (21.4)
`0 (0.0)
`17 (0.5)
`0 (0)
`0 (0.0)
`1 (0.0)
`381 (11.8)
`4 (0.1)
`803 (24.8)
`253 (7.8)
`12 (0.4)
`2 (0.1)
`170 (5.3)
`175 (5.4)
`10 (0.3)
`24 (0.7)
`
`GA20: glatiramer acetate 20 mg/mL daily, GA40: glatiramer acetate 40 mg/mL
`three-times weekly, ISR: injection-site reaction
`
`statistical significance (adjusted mean MSIS-29 physical score
`change from baseline to Month 4, 1.54 vs –0.52; treatment effect, –
`2.1; 95% CI¼–4.4 to 0.3; p¼0.090) (Supplementary Fig. 1a and b).
`As no significant treatment effect was shown in the MSIS-29
`physical score change from baseline to Month 4, subsequent sec-
`ondary endpoints were not tested for inferential differences be-
`cause of the multiplicity adjustment used in this study; associated
`p-Values for the following secondary endpoints presented in this
`paper are nominal.
`There was no difference in the MSIS-29 psychological score
`change from baseline to Month 4 between GA40 and GA20 treat-
`ment groups (adjusted mean MSIS-29 psychological score change
`from baseline to Month 4, 0.74 vs 0.02; treatment effect, –0.7; 95%
`CI¼–2.2 to 0.7; p¼0.33) (Supplementary Fig. 1a and d).
`
`3.5. Patient-reported perceptions of convenience and overall
`satisfaction using the TSQM-9
`
`An increase of more than 10 points from baseline in the un-
`adjusted TSQM-9 convenience score was already seen at Month
`1 and was sustained throughout the study for patients treated
`with GA40. However, the TSQM-9 convenience score for patients
`treated with GA20 changed only marginally throughout the study
`(Fig. 5a). Patients receiving GA40 demonstrated a nominally
`greater increase in the TSQM-9 convenience score from baseline to
`Month 4 compared with those receiving GA20 (adjusted mean
`TSQM-9 convenience score change from baseline to Month 4,
`8.8 vs 1.7; treatment effect, 7.0; 95% CI¼3.0–11.0; p¼0.0006)
`(Fig. 5b).
`There was no difference for TSQM-9 overall satisfaction score
`change from baseline to Month 4 between GA40 and GA20 treat-
`ment groups (adjusted mean TSQM-9 overall satisfaction score
`change from baseline to Month 4, 0.7 vs 1.6; treatment effect, –0.9;
`95% CI: –4.9 to 3.2; p¼0.68) (data not shown).
`
`3.6. Additional endpoints
`
`Assessment of baseline patient-reported expectations showed
`that 87% of all patients anticipated that GA40 would be more
`convenient than GA20, 86% anticipated that GA40 would be
`equally effective as GA20, and 91% expected that GA40 would be as
`safe as GA20 (Supplementary Fig. 2).
`
`4. Discussion
`
`The efficacy and safety of GA20 in RRMS is well established, as
`evidenced by multiple clinical trials and extensive post-marketing
`clinical experience (TEVA Neuroscience, Inc., 2014; Boster et al.,
`2011). Despite satisfactory efficacy, the rate of IRAEs may present a
`barrier to maintaining appropriate adherence and tolerability le-
`vels in RRMS treatment (Treadaway et al., 2009; Costello et al.,
`2008). Given that non-adherence to DMTs is associated with in-
`creased risk of relapse and poor clinical outcomes in RRMS (Tan
`et al., 2011; Al-Sabbagh et al., 2008), strategies that reduce the
`
`Fig. 5. (a) TSQM-9 convenience score. (b) Change from baseline to Month 4 in TSQM-9 convenience score. *Treatment effect (95% CI) and adjusted means (95% CI) derived
`from mixed model repeated measures analysis, adjusted for TSQM-9 subscale score at baseline, treatment group, month, and treatment by month interaction. CI: confidence
`interval, GA20: glatiramer acetate 20 mg/mL daily, GA40: glatiramer acetate 40 mg/mL three-times weekly, SE: standard error, TSQM-9: Treatment Satisfaction Questionnaire
`for Medication-9.
`
`Mylan Pharms. Inc. Exhibit 1038 Page 5
`
`

`

`J.S. Wolinsky et al. / Multiple Sclerosis and Related Disorders 4 (2015) 370–376
`
`375
`
`incidence of IRAEs and enhance patient experience are of con-
`siderable interest. Furthermore, the implementation of such ap-
`proaches in long-standing, efficacious treatments like GA, further
`enhance clinical outcomes in MS (Remington et al., 2013).
`The GLACIER study demonstrated that treatment with GA40
`was associated with a 50% reduction in the annualized rate of
`IRAEs and ISRs compared with GA20. Furthermore, a 60% reduc-
`tion in the annualized rate of moderate/severe IRAEs was observed
`with GA40 versus GA20. The frequency and event rates for the
`most common ISR types, as categorized by MedDRA 16.0 preferred
`terms, was also reduced with GA40. These safety outcomes are
`particularly notable as patients had an average of more than
`6 years of exposure to GA20 prior to the study initiation and were
`likely familiar with IRAEs. In this setting, initiation of treatment
`with GA at a higher concentration could have led to greater re-
`porting of AEs but the results presented here show that the GA40
`regimen led to a lower IRAE rate. These outcomes suggest that the
`safety profile of the new regimen has the potential to improve
`patient adherence, treatment tolerability, and clinical outcomes.
`While the observed rates of local lipoatrophy and injection-site
`necrosis were quite low with GA40 over this 4 month study and
`have been low in 1 year studies of GA 40 mg/mL used daily (Comi
`et al., 2011) or three-times weekly (Khan et al., 2013), long-term
`follow-up is required to determine if the reduced IRAE frequency
`with GA40 injections will be maintained over time.
`The GA40 dosing regimen provides the convenience of fewer
`subcutaneous injections per week than the GA20 regimen. Both
`GA40 and GA20 have been shown to be effective in reducing re-
`lapse and MRI activity compared with placebo, and both regimens
`have demonstrated a favorable safety profile (Johnson et al., 1995;
`Martinelli Boneschi et al., 2003; Comi et al., 2001; Bornstein et al.,
`1987; Khan et al., 2013). In GLACIER, at baseline, 87% of enrolled
`patients expected GA40 to be more convenient than GA20. Fol-
`lowing the open-label, head-to-head treatment assignment, pa-
`tients' expectations that GA40 injections three-times weekly would
`be more convenient than daily injections of GA20 were confirmed.
`At baseline, 86% of GLACIER participants expected that GA40
`and GA20 would have similar effectiveness. Indeed, GA20 and
`GA40 had similar effects on patient-reported measures of physical
`and psychological well-being using the MSIS-29. While changes in
`MSIS-29 scores were similar between treatment regimens, the
`treatment advantage of GA40 over GA20 is best represented by
`safety, tolerability, and convenience outcomes.
`Because investigators and participants were neither blinded
`nor masked to the treatment assignment, GLACIER study outcomes
`should be interpreted in the context of this acknowledged study
`limitation. However, these trial design features were intentional,
`so as to recapitulate actual clinical and patient experiences.
`
`written informed consent before any study-related procedures
`were performed.
`
`Declaration of conflict of interest
`
`Dr. Wolinsky has received compensation for serving on steering
`committees or data monitoring boards for Novartis, Roche, Sanofi,
`and Teva; reimbursement for services as a consultant to AbbVie,
`Acetilon, Alkermes, Athersys, EMD Serono, Genzyme, Novartis,
`Roche, Teva, and XenoPort; royalty payments through the Uni-
`versity of Texas Health Science Center at Houston for monoclonal
`antibodies out-licensed to Chemicon International; and research
`support from Genzyme, Sanofi, and the National Institutes of
`Health through the University of Texas Health Science Center at
`Houston.
`Dr. Borresen has received compensation from Roche, Novartis,
`and Teva.
`Dr. Dietrich has received compensation for speaking for Biogen
`IDEC, Novartis, and Teva and performing advisory board services
`for Teva.
`Dr. Wynn has received compensation for speaking and/or
`consulting for Acorda Therapeutics, Avanir Pharmaceuticals, EMD
`Serono, Genzyme, GlaxoSmithKline, Sanofi, Teva, and Xenoport;
`and research support from Acorda Therapeutics, Avanir Pharma-
`ceuticals, EMD Serono, Genentech/Hoffman LaRoche, Sanofi/Gen-
`zyme, GlaxoSmithKline, Novartis, Osmotica, Xenoport, Teva, and
`the National MS Society.
`Drs. Sidi, Steinerman, Knappertz, and Kolodny are employees of
`Teva Pharmaceutical Industries.
`
`Funding
`
`Trial NCT01874145 was funded by Teva Pharmaceutical In-
`dustries Ltd, Petach Tikva, Israel.
`
`Acknowledgments
`
`We thank the patients and site personnel involved with this
`study, Robin Everts and Svetlana Rubinchick (of Teva Pharmaceu-
`ticals) for assistance with study conduct and statistical analyzes,
`and Peter Feldman, PhD (Teva Pharmaceuticals) and Rhonda
`Charles, PhD (Chameleon Communications International with
`funding from Teva Pharmaceutical Industries Ltd) for editorial
`assistance in the preparation of this report. This study was funded
`by Teva Pharmaceutical Industries Ltd, Petach Tikva, Israel.
`
`5. Conclusions
`
`Appendix A. Supplementary material
`
`The GLACIER study provides open-label, randomized, parallel-
`arm, prospective, comparative data demonstrating that GA40 has a
`favorable safety profile with significant improvement in the rate of
`IRAEs, and an en