Glatiramer acetate recovers microscopic tissue damage in patients with multiple sclerosis. A case control diffusion imaging study

Transcription

1 Pathophysiology 18 (2011) Glatiramer acetate recovers microscopic tissue damage in patients with multiple sclerosis. A case control diffusion imaging study R. Zivadinov a,b,, Sara Hussein a, Milena Stosic a, Jacqueline Durfee a, Jennifer L. Cox a, Diane L. Cookfair a,b, Komal Hashmi a, Nadir Abdelrahman a, Neeta Garg b, Michael G. Dwyer a, B. Weinstock-Guttman b a Buffalo Neuroimaging Analysis Center, Department of Neurology, University of Buffalo, Buffalo, NY, United States b The Jacobs Neurological Institute, Department of Neurology, University of Buffalo, Buffalo, NY, United States Received 27 February 2010; received in revised form 16 March 2010; accepted 8 April 2010 Abstract Traditional magnetic resonance imaging (MRI) techniques have contributed to the management of multiple sclerosis (MS) but are limited in their ability to detect neuronal damage. Advanced MRI metrics provide assessment of microscopic neuronal changes; however, few studies have examined the effects of MS therapies on these measures. This prospective, open-label, observational study evaluated the effect of subcutaneous glatiramer acetate (GA) 20 mg/day on the 1- and 2-year changes in diffusion-weighted imaging (DWI) measures in patients with relapsing remitting (RR) MS and in age- and sex-matched healthy controls (HC). Inclusion criteria were age 18 65, RR disease course, expanded disability status scale (EDSS) score 5.5 and disease duration < 20 years. MS patients and HC underwent 1.5 T MRI scans and clinical examinations at baseline and at 1- and 2-year follow-up. Nineteen RRMS patients and 16 HC completed the 1-year follow-up and 16 MS patients and 13 HC the 2-year follow-up of the study. In MS patients, treatment with GA promoted recovery of DWI mean parenchymal diffusivity (MPD) at year 1 ( 7.1%, p = 0.007) and at year 2 ( 10.1%, p = 0.028). The recovery of DWI MPD was significantly higher in MS patients compared to HC at year 1 (p = 0.01) and year 2 (p < 0.001). GA promoted recovery of DWI entropy at 2 years ( 1.2%, p = 0.018). No significant DWI MPD and entropy changes were observed in HC over the follow-up. No significant deterioration in magnetization transfer ratio occurred over the follow-up in MS patients and HC. Patients on GA and HC did not develop significant global or regional atrophy over 2 years. GA significantly improved microscopic tissue damage in the brain, as measured by DWI over the 1- and 2-year follow-up Elsevier Ireland Ltd. All rights reserved. Keywords: Multiple sclerosis; Diffusion-weighted imaging; Glatiramer acetate; Magnetization transfer imaging; Brain atrophy 1. Introduction Multiple sclerosis (MS) is a central nervous system (CNS) autoimmune disorder characterized by chronic inflammation, axonal demyelination, and neuronal degeneration [1]. The complex pathophysiology of MS is mediated by autoreactive T cells that ultimately invade and attack the CNS by traversing the blood brain barrier, resulting in persistent irreversible axonal damage and physical disability [2]. Magnetic Corresponding author at: The Jacobs Neurological Institute, Department of Neurology, University of Buffalo, School of Medicine and Biomedical Sciences, 100 High Street, Buffalo, NY 14203, United States. Tel.: ; fax: address: rzivadinov@bnac.net (R. Zivadinov). resonance imaging (MRI) metrics contribute substantially to the diagnosis and management of MS, providing critical information regarding relapse activity and disease severity in addition to serving as secondary endpoint measures for clinical trials evaluating MS therapies [3,4]. Traditional MRI techniques (e.g., gadolinium (Gd)-enhanced T1-weighted imaging, spin-echo [SE] T2-weighted imaging [T2-WI]) have proven effective for detecting the presence of macroscopic white matter MS lesions but are limited by the inability to effectively delineate microscopic tissue damage occurring in normal-appearing white matter (NAWM) and microscopic and macroscopic tissue damage in normal-appearing gray matter (NAGM) [5,6]. Furthermore, the brain-lesion burden on traditional MRI scans correlates only modestly with clinical disease progression [7]. Thus, the need exists for /$ see front matter 2010 Elsevier Ireland Ltd. All rights reserved. doi: /j.pathophys

2 62 R. Zivadinov et al. / Pathophysiology 18 (2011) alternative MRI techniques that more accurately monitor clinical disease activity and efficacy of disease-modifying therapies. In this regard, advanced MRI measures (e.g., diffusionweighted imaging [DWI], magnetization transfer imaging [MTI], and magnetic resonance spectroscopy [MRS]) are emerging that show greater sensitivity to pathologic changes occurring in patients with MS, including metabolic changes and neuronal tissue damage [8]. These techniques appear to be reliable surrogate markers of the microscopic pathologic processes contributing to disease activity and progression, including demyelination, axonal loss, and neurodegeneration [6]. Diffusion-weighted imaging is a unique, advanced technique that provides quantitative assessment of the microscopic disease burden in brain tissue not visible with traditional MRI techniques. Specifically, DWI evaluates the effect of the cellular environment of CNS tissue on the diffusion properties of water molecules, and measurements obtained from DWI scans provide information regarding changes in the size, shape, geometry, and orientation of neuronal tissue [9]. Demyelination and neurodegeneration can alter the geometry of brain tissue orientation, resulting in an increase in water diffusivity measurable with DWI indices [10]. Information provided by DWI scans could substantially benefit patients with MS because of the potential ability to identify disease activity in regions of the CNS lacking overtly active disease lesions (i.e., NAWM, NAGM) and to detect microscopic alterations in neuronal tissue (e.g., regions of demyelination or inflammation) that substantially precede the development of overt lesions visible on traditional MRI scans [11]. In this regard, DWI could potentially function as a predictor of neuronal damage and physical disability. Because advanced MRI metrics provide a more quantitative and sensitive approach to understanding the dynamic nature of MS pathophysiology, they potentially allow a more effective clinical evaluation of MS therapies [12]. Although an increasing number of clinical studies have applied advanced metrics to the evaluation of neuronal damage in patients with MS [10], only a few studies have evaluated the effects of disease-modifying MS therapies (e.g., glatiramer acetate [GA], interferon [IFN] beta-1a, and IFN beta-1b) on MTI measures [13]. To the best of our knowledge, no studies have investigated the effect of these therapies on DWI measures [13,14]. Glatiramer acetate is a synthetic copolymer with a favorable safety and efficacy profile in the treatment of relapsing remitting MS (RRMS) [15]. Glatiramer acetate is thought to exhibit its therapeutic effect by inducing the production of anti-inflammatory T cells that cross the blood brain barrier, increase secretion of inhibitory cytokines, and downregulate inflammatory activity within the CNS [16]. Additionally, in a preclinical study, GA-specific autoreactive T cells have been shown to provide neuroprotection of the optic nerve [17]. However, based on clinical assessments applying traditional MRI measures, it is unclear whether GA could prevent pathological changes at the cellular level. Thus, the overall goal of this study was to investigate the effect of GA on traditional and advanced MRI measures after 2 years of treatment in patients with RRMS and ageand sex-matched healthy controls (HC). 2. Methods 2.1. Study design and treatment This was a phase 4, prospective, open-label, singleblinded, observational study involving 19 patients with RRMS followed at the Jacobs Neurological Institute, University at Buffalo, Buffalo, NY, who were treated with Copaxone monotherapy (20 mg/day sc) for a mean period of 2.3 years and of 16 sex- and age-matched normal controls (HC). Eligible patients (18 65 years) had a clinically definite MS diagnosis based on the McDonald criteria [18], a Kurtzke expanded disability status scale (EDSS) score 6.5 [19], a disease duration of 6 months to 20 years, and had received treatment with GA for a minimum of 6 months before study entry. During the 3 months before study entry, patients could not have received treatment with any immunosuppressant agents (e.g., mitoxantrone, cyclophosphamide, cladribine, fludarabine, cyclosporine), total body irradiation, steroid therapy for symptom relapses, or any other concomitant immunomodulatory therapies (e.g., azathioprine, methotrexate, intravenous immunoglobulin, mycophenolate mofetil, natalizumab, IFN beta) except GA. A physical exam and neurologic assessment, including EDSS score, were administered at baseline, 12 and 24 months, and within 5 days of relapse if symptoms occurred. Blood tests for safety monitoring were performed at baseline, 12 and 24 months. Eligible patients received monotherapy with subcutaneous injections of GA 20 mg once daily for 24 months. In the event of a symptom relapse, intravenous methylprednisolone 1 g was co-administered daily for 3 5 days. The study was approved by local Ethical Committee MRI assessments MRI acquisition and analysis MRI scans were performed at baseline, 12 and 24 months using a 1.5-Tesla General Electric Signal unit (Milwaukee, Wisconsin). For each session, the following images were acquired, applying the same protocol at all time points: DWI; T2-WI; three-dimensional (3D) spoiled gradient-recalled (SPGR) T1-WI; SE T1-WI with and without Gd contrast; fast fluid-attenuated inversion recovery (FLAIR); proton density (PD); and PD with magnetization transfer (MT) pulse images. The Gd-enhanced SE T1-WI sequence was acquired after injection of a single intravenous bolus dose (0.1 mmol/kg Gd-diethylenetriaminepentaacetic acid) within 5 min after administration of contrast agent. Echo-planar imaging was used to acquire T2-weighted (T2-W; B 0 ) and diffusion-weighted (DW; B 1000 ) images. Diffusion weighting was applied via identical diffusion-encoding waveforms

3 R. Zivadinov et al. / Pathophysiology 18 (2011) in 3 orthogonal directions (read, phase-encoding, and sliceselection), resulting in an average DW image with a b-factor of 1000 s/mm 2. Study patients and HC were positioned in the MRI unit according to commonly accepted international guidelines [20]. Image analyses were blinded to patients clinical characteristics and clinical status. At all stages of automated analysis, quality control montage imaging output analysis files were examined by an expert observer to verify optimal performance of the automated algorithms Lesion and atrophy measures. Assessments of T2, T1, and Gd lesion volumes (LV) were performed using a semi-automated edge detection contouring-thresholding technique described previously [21]. For brain extraction and tissue segmentation, longitudinal and cross-sectional brain atrophy analysis software (SIENA and SIENAX, respectively) was employed as previously described [22 25]. Compartment-specific absolute volumes, including partial volume estimates, were quantified and multiplied by the patient-specific normalization factor to obtain normalized tissue volume measures. Normalized brain volume (NBV), normalized gray matter volume (NGMV), and normalized white matter volume (NWMV) were obtained at baseline. To compare tissue compartments longitudinally, cross-sectional SIENAX analysis was performed and absolute compartment-wise differences were applied to calculate mean percent compartment volume changes. An analysis of relative edge motion was used to calculate the percent brain volume change (PBVC) between the two time intervals MTI measures. Magnetization transfer imaging was performed by completely automated software [26,27]. Initially, FLAIR, 3D-SPGR, T1-WI, and PD-MT images were co-registered to a native acquisition space PD image [28] and SIENAX segmentation of the native space 3D-T1-SPGR was performed [25]. Similarly, user-generated FLAIR and T1-WI lesion masks were co-registered with the PD image based on their transformation matrices. Next, the realigned segmented SPGR was used to create separate parenchyma and cerebrospinal fluid (CSF) masks in PD space. The CSF mask was radially expanded outward by 1 pixel to create an eroded parenchymal mask, preventing erroneous inclusion of CSF. The parenchymal mask was then applied to the PD image, removing all extra-parenchymal structures as well as internal CSF. The FLAIR and T1-WI lesion masks were used to nullify overt lesions, and the remaining PD was algebraically combined with the co-registered PD-positive MT image to generate a whole brain (WB) map, applying the equation: PD PD + MT %MTR = 100 PD The mean magnetization transfer ratio (MTR), histogram peak position, and histogram peak height of the final image were calculated using only pixels of intensity >10 pixel units. The scan rescan reproducibility was as follows: 0.27% for MTR WB, 0.77% for T2 lesion volume and 0.9% for T1 lesion volume. The scan rescan reproducibility was higher for peak height and peak position MTR measures (range 0.9 4%). Given the superior reliability of mean MTR and to minimize the number of multiple comparisons, the analyses emphasized the mean MTR measures DWI measures. A fully automated technique was employed to obtain DWI scans [29,30]. Briefly, echo-planar T2-WI and DW images were combined to create a mean diffusivity ( D) map by performing the following calculation on a voxel-by-voxel basis: D = ln DW/T2 b where DW represents the average diffusion signal intensity and T2 represents the T2-W signal intensity without diffusion weighting. The T2-WI was then subjected to a number of algorithms to extract the brain parenchyma [25,31], leaving a parenchymal mask image. The parenchymal mask image was applied to the previously created ( D) map to create a parenchymal ( D) map, displaying mean parenchymal diffusivity (MPD) values for the parenchyma only. To avoid contamination by partial voluming of CSF, the resulting parenchymal map was eroded by 1 voxel using a 3D erosion algorithm to yield only those voxels completely within the brain tissue. Histogram analysis for each patient was then performed on the eroded parenchymal map, using 1% of the maximum value as the bin width. All histograms were normalized so that the sum of all voxel counts was 1. Measures of MPD and entropy were obtained. Additionally, histogram entropy was calculated as follows: Entropy = Σ( p i )[log(p i )] where p i represents the probability of finding intensity i in the image, calculated as follows: number of voxels (i) p i = number of voxels (entire image) and is equivalent to the peak height of intensity i, because frequencies are normalized to sum to unity [30]. The scan rescan reproducibility was 0.84% for MPD and 0.61% for entropy Study endpoints The primary endpoint of the study was the percent change from baseline in DWI MPD and entropy at 12 and 24 months. Secondary endpoints included the following measures: percent change from baseline in brain T2 and T1 lesion MTR and WB MTR at 12 and 24 months; brain atrophy measures; percent change from baseline of Gd-, T2- and T1-LV at 12 and 24 months and disability progression at 12 and 24 months (i.e., 1-point worsening from baseline for EDSS score of 5.5 or 0.5-point worsening from baseline for EDSS score of 6); and relapse-related variables (e.g., total number of relapses, relapse-free). Symptom relapse was defined as the appearance or reappearance of 1 symptom attributable to

4 64 R. Zivadinov et al. / Pathophysiology 18 (2011) MS, accompanied by objective deterioration on neurologic examination, lasting at least h, present without fever, and preceded by neurologic stability for at least 30 days and without steroid withdrawal within 60 days of the new event Statistical analysis Statistical analysis was performed using the Statistical Package for the Social Sciences (SPSS, Version 13.0). For comparisons between the RRMS and HC groups, the chisquare test was applied for non-continuous data and the Mann Whitney U test was employed for between-group comparisons involving continuous, non-normally distributed data. The effect of subcutaneous GA treatment on outcome measures was determined by comparing changes from baseline to follow-up for each patient. Within-subject changes from baseline MRI measures were tested using the Wilcoxon signed rank test for paired data. The minimum significance level for entry and for staying in the equation was All p values were based on 2-tailed tests. 3. Results 3.1. Demographic and clinical characteristics Of the 19 RRMS patients enrolled in the study, all 19 completed the 1-year follow-up and 16 patients completed the 2-year follow-up. Of 16 HC enrolled in the study, all 16 completed the 1-year follow-up and 13 completed the 2-year follow-up. One patient refused to participate due to neoplasm; 1 patient completed the clinical 2-year follow-up but interrupted the MRI scan during acquisition and did not want to repeat it; 1 patient refused to participate in the 2- year follow-up study. All 3 HC refused to participate in the 2-year follow-up study. The baseline characteristics of the entire study cohort are shown in Table 1. There were no significant differences at baseline for the demographic, clinical and MRI characteristics between the RRMS and HC for the 1- and 2-year study cohorts MRI outcomes Baseline MRI characteristics in MS patients and HC are shown in Table 2. Treatment with GA promoted recovery of DWI MPD at 1 year ( 7.1%, p = 0.007) and at 2 years ( 10.1%, p = 0.028) in MS patients (Table 3 and Fig. 1). The recovery of DWI MPD was significantly higher in MS patients compared to HC at year 1 (p = 0.01) and year 2 (p < 0.001). Treatment with GA promoted recovery of DWI entropy at 2 years ( 1.2%, p = 0.018) in MS patients (Table 3 and Fig. 1). No differences were observed between MS patients and HC for DWI entropy, although there was a trend for higher decrease in entropy over 2 years in the MS group (p = 0.09). No significant DWI MD and entropy changes were observed within HC over the follow-up. Table 1 Baseline demographic and clinical characteristics in MS patients and healthy controls. Relapsing remitting (n = 19) Healthy controls (n = 16) p value % Female NA Age, years, mean (SD) 42.3 (11.2) 44.2 (5) NA Age onset, years, mean 32.7 (11.9) NA NA (SD) Disease duration, years, 9.5 (8.3) NA NA mean (SD) Relapses in 1 year 0.43 (0.8) NA NA before study baseline, mean (SD) EDSS, mean (SD) 2.3 (1.5) NA NA Number of relapses in 1 year before study baseline (%) NA NA >3 5.3 n: number; SD: standard deviation; NA: not available; EDSS: expanded disability status scale. No significant within-subject deterioration in MTR measures of WB, T2-LV and T1-LV was observed in MS patients and HC over follow-up (Table 3). There were no significant differences between MS patients and HC in WB MTR changes over follow-up. Patients on GA and HC did not develop significant global or regional atrophy over 2 years (Table 3). In particular, PBVC declined only 0.8% over 2 years and GMV declined 1.2%, respectively (p = NS) (Table 3). However, the PBVC decline was higher in MS patients compared to controls over 2 years (p = 0.01, Table 3). Table 2 Baseline MRI characteristics in MS patients and healthy controls. RRMS group (n = 19) HC group (n = 16) p value Gd positivity status, % 63.2 NA NA Gd lesion number 1.9 ± 2.6 NA NA Gd lesion volume, ml 0.1 ± 0.3 NA NA T2 lesion volume, ml 7.3 ± 7.5 NA NA T1 lesion volume, ml 0.7 ± 0.9 NA NA NBV, ml ± ± 37.7 < NGMV, ml ± ± NWMV, ml ± ± DWI MPD, 10 6 /mm 2 /s ± ± DWI entropy 3.2 ± ± 0.1 < MTR T2 lesion volume 35.8 ± 3.1 NA NA MTR T1 black hole 30.6 ± 3.0 NA NA lesion volume MTR WB 37.1 ± ± n: number; MS: multiple sclerosis; RR: relapsing remitting; HC: healthy control; Gd: gadolinium; NA: not available; LV-lesion volume, NBV: normalized brain volume; NGMV: normalized gray matter volume; NWMV: normalized white matter volume; DWI: diffusion-weighted imaging; MPD: mean parenchymal diffusivity; MTR: magnetization transfer ratio; WB: whole brain.

5 R. Zivadinov et al. / Pathophysiology 18 (2011) Table 3 Within-subject percent MRI changes in conventional and non-conventional MRI measures over follow-up period in MS patients and healthy controls. MRI changes Baseline to 1 year change in MS patients (n = 19) mean % (SD) Baseline to 2 years change in MS patients (n = 16) mean % (SD) Baseline to 1 year change in HC (n = 16) mean % (SD) Gadolinium status NA NA positivity at follow-up (%) Gadolinium LV 95 (0.1) *** 100 (0) *** NA NA T2-LV +12 (65) (48) NA NA T1-LV 22 (48.6) 38.2 (26.4) NA NA PBVC 0.45 (2) 0.8 (0.6) 0.21 (0.05) 0.38 (0.07) GMV 0.7 (3.8) 1.2 (1.9) 0.35 (0.1) 0.58 (0.2) WMV +1.9 (5.7) +1.8 (2.6) +0.1 (0.1) +0.2 (0.1) DWI MPD 7.1 (11.1) ** 10.1 (5.1) * +0.4 (1) +0.7 (1) DWI entropy +1.1 (11.3) 1.2 (2.8) * +0.1 (0.05) (0.05) MTR T2-LV 0.7 (13) 3.3 (3.9) NA NA MTR T1-LV +0.9 (19) 5.4 (8.8) NA NA MTR WB +0.7 (4.4) 1.4 (2.8) 0.3 (2.1) 0.5 (1.2) Baseline to 2 years change in HC (n = 13) mean % (SD) Mann Whitney Rank Sum Test was used to detect significant differences between MRI changes in MS and HC over 1 year and 2 years. The following significant differences were found over 1 year (higher MPD decrease in MS p = 0.01) and over 2 years (higher PBVC decrease in MS p = 0.01, higher WMV increase in MS p = 0.03 and higher MPD decrease in MS p < ). n: number; MS: multiple sclerosis; HC: healthy control; SD: standard deviation; LV: lesion volume; NA: not available; PBVC: percent brain volume change; GMV: gray matter volume; WMV: white matter volume; DWI: diffusion-weighted imaging; MPD: mean parenchymal diffusivity; MTR: magnetization transfer ratio; WB: whole brain. * Wilcoxon Rank Sum test was used to detect significant changes within groups (<0.05). ** Wilcoxon Rank Sum test was used to detect significant changes within groups (<0.01). *** Wilcoxon Rank Sum test was used to detect significant changes within groups (<0.001). Over the 1- and 2-year follow-up, there was a significant decrease in Gd-enhancement activity (p < ) (Table 3) Clinical outcomes Overall, RRMS patients were stable in their disability status over the follow-up period, with only 2 patients (12.5%) exhibiting a 1-point increase in EDSS score over the 2- year follow-up. Furthermore, 11 patients (69%) experienced Fig. 1. Percent within-subject change from baseline in diffusion-weighted imaging (DWI) measures after 12 and 24 months of glatiramer acetate therapy in MS patients and healthy controls. MPD: mean parenchymal diffusivity; RRMS: relapsing remitting multiple sclerosis; HC: healthy controls. no symptom relapse, and only 1 patient (12.5%) experienced 1 relapse over 1-year follow-up and 5 patients (31.3%) over 2-year follow-up. 4. Discussion Traditional MRI techniques (e.g., Gd-enhanced T1-WI, SE T2-WI) are effective diagnostic tools for MS but do not consistently correlate with disease progression and are limited by their ability to detect only macroscopic, overt lesions of MS disease activity [7,11]. Although no specific, highly predictive marker of clinical outcome for patients with MS has been identified, a number of advanced MRI measures are emerging with greater sensitivity for measuring pathologic processes associated with MS disease activity and progression, including the ability to assess microscopic areas of disease pathology in the NAGM and NAWM [8]. These advanced MRI metrics provide a more quantitative, sensitive approach to understanding the dynamic nature of MS pathophysiology and potentially allow for a more effective clinical evaluation of MS therapies. Recent studies have demonstrated the validity of a novel, fully automated method for the calculation of quantitative DWI measures. One clinical study assessed the correlation between WB DWI variables and MRI measures of disease severity in a large cohort of consecutive patients with MS (n = 432) evaluated by 1.5 T brain MRI [30]. The authors reported the validity of DWI in discerning differ-

6 66 R. Zivadinov et al. / Pathophysiology 18 (2011) ences between patients with MS and healthy controls, as well as between different subtypes of patients with MS. Both DWI measures of MPD and entropy were sensitive markers of overall brain damage and strongly related to clinical impairment in patients with MS. An additional clinical study of 60 patients with MS (RRMS, n = 40; SPMS, n = 20) investigated the correlation between DWI measures and cognitive dysfunction in patients with MS [29]. Neuropsychologic examinations evaluated auditory/verbal and visual/spatial memory and assessed processing speed and executive function. Substantial correlations were observed between DWI and neuropsychologic performance in all cognitive domains. Therefore, given its overall simplicity and strong correlation with physical disability and cognition, DWI may offer several logistic advantages over more traditional MRI measures when predicting clinical impairment. In the present study, the effect of GA on DWI measures (e.g., MPD and entropy) was particularly robust, and the recovery in DWI measures following GA therapy may explain the relative stability of clinical status over 2 years in patients with RRMS. This prospective study is one of the first to examine the effects of GA on DWI measures in patients with MS. Glatiramer acetate has demonstrated effectiveness in the treatment of RRMS and it has been suggested that it confers neuroprotective effects in the CNS through a mechanism of action involving the release of anti-inflammatory cytokines and neurotrophic factors by Th2 cells [15,16]. However, based on clinical assessments applying traditional MRI measures, it remains unclear whether GA can actually prevent pathological changes at the cellular level. Findings from the present study demonstrate that GA treatment improved or stabilized disease progression on advanced MRI assessments of DWI, MTR and brain atrophy. Overall, only 12.5% of patients experienced a 1-point increase in EDSS score, and 69% of patients reported no symptom relapse. These findings suggest that GA might have the ability to prevent both macroscopic and microscopic tissue damage within the CNS. MS is a chronic progressive disease of variable course in which long-term benefits of treatment are critically important. Different immunomodulatory treatments have been approved for the treatment of MS, and all have demonstrated short-term effectiveness in reducing symptom relapse rates, MRI measures of disease activity, and disease progression [32]. The effect of GA has been studied prospectively over a period of 10 years and outcomes have been compared among patients continuing GA treatment and those who have discontinued GA therapy. A clinical report of 10-year data showed stabilized suppression of relapse rates, with 62% of patients experiencing stable or improved disability levels disease progression rates considerably less than expected from natural history studies of untreated patients with MS [33]. The present study included a well-defined cohort of patients with RRMS who received GA monotherapy for an average of 4.3 years. Although based on a small number of patients evaluated for a relatively short period of time, results from this study suggest that GA may exert a protective effect on MS disease pathology, mainly by reducing cellular damage, as measured by DWI. A small number of studies have evaluated the effects of immunomodulatory treatment on MTR measures and demonstrated potential improvement in lesional MTR but no significant effect on WB MTR [13,14,34,35]. In one clinical study of RRMS patients (n = 8), treatment with IFN -1b substantially reduced the number of Gd-positive lesions but had no effect on WB MTR over a 6-month period [35]. In another clinical study of 4 patients with RRMS, IFN -1b treatment did not alter MTR reduction in areas that subsequently developed Gd-positive lesions [14]. Furthermore, a randomized, placebo-controlled study of long-term IFN -1b therapy (i.e., 1, 2, or 3 years) reported no effect of IFN -1b therapy in patients with MS (n = 42), compared with placebo (n = 40), on WB MTR measures [34]. In this regard, no significant deterioration in MTI measures was observed in the present study. These findings are in agreement with previously published data reporting the inhibited progression of Gd-enhancing lesions to chronic T1 hypointense black holes in response to GA therapy [36]. Finally, a placebo-controlled clinical study demonstrated that patients who received GA therapy (n = 102) experienced slowed progression of brain atrophy compared with patients who received placebo (n = 105) [37]. Patients treated with GA experienced decreases in PBVC of 0.8% during the first 9 months of the study, 0.6% during the second 9 months, and 1.5% during the total 18-month period. In the present study, patients treated with GA decreased only 0.45% in PBVC over the first year and 0.8% over the period of 2 years. The present study has some limitations that warrant comment. First, this open-label, post-marketing study involved a small number of patients, and multiple tests were performed as part of this study, so results must be interpreted with caution. Additionally, although the study included HC, it lacked a patient control group and, although patients were enrolled on a consecutive basis, they were not randomized or followed in a blinded fashion, except for MRI analysis. Although the study patients exhibited low overall clinical disease activity, this preliminary study included a well defined cohort of patients with RRMS who received GA monotherapy for an average of 2.3 years before study entry. Results of the two a priori primary endpoints suggest that GA may provide multiple clinical benefits for patients with MS, including a slowed progression of advanced MRI measures. However, more long-term clinical follow-up is warranted. The extensive MRI protocol performed in this study could be improved with the addition of other advanced techniques, including MRS and functional MRI measures. This would provide an opportunity to better understand the demyelination, remyelination, neurodegeneration, and regeneration processes occurring as a result of long-term GA therapy in patients with MS. In summary, the preliminary findings from the present study demonstrate that long-term treatment with GA slows disease progression on both traditional and advanced MRI

7 R. Zivadinov et al. / Pathophysiology 18 (2011) measures; the findings also suggest that GA may prevent neuronal damage in patients with MS. Acknowledgments This study was funded by an investigator-initiated grant awarded to R. Zivadinov, MD, PhD, from Teva Neuroscience, Kansas City, Missouri. The authors thank Eve Salczynski for technical assistance in the preparation of this manuscript. References [1] R.J. Fox, F. Bethoux, M.D. Goldman, J.A. Cohen, Multiple sclerosis: advances in understanding, diagnosing, and treating the underlying disease, Cleve. Clin. J. Med. 73 (2006) [2] R. Martin, H.F. McFarland, D.E. McFarlin, Immunological aspects of demyelinating diseases, Annu. Rev. Immunol. 10 (1992) [3] R. Bakshi, A. Minagar, Z. Jaisani, J.S. Wolinsky, Imaging of multiple sclerosis: role in neurotherapeutics, NeuroRx 2 (2005) [4] M. Filippi, A. Falini, D.L. Arnold, F. Fazekas, O. Gonen, J.H. Simon, V. Dousset, M. Savoiardo, J.S. Wolinsky, Magnetic resonance techniques for the in vivo assessment of multiple sclerosis pathology: consensus report of the white matter study group, J. Magn. Reson. Imaging 21 (2005) [5] R. Zivadinov, M. Stosic, J.L. Cox, D.P. Ramasamy, M.G. Dwyer, The place of conventional MRI and newly emerging MRI techniques in monitoring the different aspects of treatment outcomes, J. Neurol. 255 (Suppl. 1) (2008) [6] R. Zivadinov, Evidence for neuroprotection in multiple sclerosis: can imaging techniques measure neuroprotection and remyelination? Neurology 68 (Suppl. 3) (2007) S72 S82. [7] R. Zivadinov, T.P. Leist, Clinical-magnetic resonance imaging correlations in multiple sclerosis, J. Neuroimaging 15 (Suppl. 4) (2005) 10S 21S. [8] R. Bakshi, Magnetic resonance imaging advances in multiple sclerosis, J. Neuroimaging 15 (Suppl. 4) (2005) 5S 9S. [9] C. Pierpaoli, P. Jezzard, P.J. Basser, A. Barnett, G. Di Chiro, Diffusion tensor MR imaging of the human brain, Radiology 201 (1996) [10] M. Rovaris, A. Gass, R. Bammer, S.J. Hickman, O. Ciccarelli, D.H. Miller, M. Filippi, Diffusion MRI in multiple sclerosis, Neurology 65 (2005) [11] D.H. Miller, A.J. Thompson, M. Filippi, Magnetic resonance studies of abnormalities in the normal appearing white matter and grey matter in multiple sclerosis, J. Neurol. 250 (2003) [12] D.A. Carone, R. Zivadinov, B. Weinstock-Guttman, J. Gaines, C. Maggiore, J. Sharma, M. Tomassi, R. Bakshi, R. Benedict, Learning inconsistency is associated with frontal lobe atrophy in multiple sclerosis (MS), J. Int. Neuropsychol. Soc. 11 (2005) 42. [13] R. Zivadinov, Advanced magnetic resonance imaging metrics: Implication in multiple sclerosis clinical trials, Meth. Find Exp. Clin. Pharmacol. 31 (2009) [14] P.E. Sijens, J.P. Mostert, R. Irwan, J.H. Potze, M. Oudkerk, J. De Keyser, Impact of fluoxetine on the human brain in multiple sclerosis as quantified by proton magnetic resonance spectroscopy and diffusion tensor imaging, Psychiatry Res. 164 (2008) [15] K.P. Johnson, B.R. Brooks, J.A. Cohen, C.C. Ford, J. Goldstein, R.P. Lisak, L.W. Myers, H.S. Panitch, J.W. Rose, R.B. Schiffer, T. Vollmer, L.P. Weiner, J.S. Wolinsky, The Copolymer 1 Multiple Sclerosis Study Group, 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 45 (1995) [16] V.W. Yong, Differential mechanisms of action of interferon-beta and glatiramer acetate in MS, Neurology 59 (2002) [17] J. Kipnis, E. Yoles, Z. Porat, A. Cohen, F. Mor, M. Sela, I.R. Cohen, M. Schwartz, T cell immunity to copolymer 1 confers neuroprotection on the damaged optic nerve: possible therapy for optic neuropathies, Proc. Natl. Acad. Sci. U.S.A. 97 (2000) [18] W.I. McDonald, A. Compston, G. Edan, D. Goodkin, H.P. Hartung, F.D. Lublin, H.F. McFarland, D.W. Paty, C.H. Polman, S.C. Reingold, M. Sandberg-Wollheim, W. Sibley, A. Thompson, S. van den Noort, B.Y. Weinshenker, J.S. Wolinsky, Recommended diagnostic criteria for multiple sclerosis: guidelines from the International Panel on the diagnosis of multiple sclerosis, Ann. Neurol. 50 (2001) [19] J.F. Kurtzke, Rating neurologic impairment in multiple sclerosis: an expanded disability status scale (EDSS), Neurology 33 (1983) [20] D.H. Miller, F. Barkhof, I. Berry, L. Kappos, G. Scotti, A.J. Thompson, Magnetic resonance imaging in monitoring the treatment of multiple sclerosis: concerted action guidelines, J. Neurol. Neurosurg. Psychiatry 54 (1991) [21] R. Zivadinov, R.A. Rudick, R. De Masi, D. Nasuelli, M. Ukmar, R.S. Pozzi-Mucelli, A. Grop, G. Cazzato, M. Zorzon, Effects of IV methylprednisolone on brain atrophy in relapsing remitting MS, Neurology 57 (2001) [22] R.H. Benedict, J.M. Bruce, M.G. Dwyer, N. Abdelrahman, R. Zivadinov, Neocortical atrophy, third ventricular width, and cognitive dysfunction in multiple sclerosis, Arch. Neurol. 53 (2006) [23] R. Zivadinov, B. Weinstock-Guttman, R. Benedict, M. Tamano-Blanco, S. Hussein, N. Abdelrahman, J. Durfee, M. Ramanathan, Preservation of gray matter volume in multiple sclerosis patients with the Met Allele of the rs6265 (Val66Met) SNP of brain derived neurotrophic factor (BDNF), Hum. Mol. Genet. 16 (2007) [24] S.M. Smith, N. De Stefano, M. Jenkinson, P.M. Matthews, Normalized accurate measurement of longitudinal brain change, J. Comput. Assist. Tomogr. 25 (2001) [25] S.M. Smith, Y. Zhang, M. Jenkinson, J. Chen, P.M. Matthews, A. Federico, N. DeStefano, Accurate, robust, and automated longitudinal and cross-sectional brain change analysis, Neuroimage 17 (2002) [26] B. Weinstock-Guttman, R. Zivadinov, M. Tamano-Blanco, N. Abdelrahman, D. Badgett, J. Durfee, S. Hussein, J. Feichter, K. Patrick, R. Benedict, M. Ramanathan, Immune cell BDNF secretion is associated with white matter volume in multiple sclerosis, J. Neuroimmunol. 188 (2007) [27] C.W. Tjoa, R.H. Benedict, M.G. Dwyer, D.A. Carone, R. Zivadinov, Regional specificity of magnetization transfer imaging in multiple sclerosis, J. Neuroimaging 18 (2008) [28] M. Jenkinson, P. Bannister, M. Brady, S. Smith, Improved optimization for the robust and accurate linear registration and motion correction of brain images, Neuroimage 17 (2002) [29] R.H. Benedict, J. Bruce, M.G. Dwyer, B. Weinstock-Guttman, C. Tjoa, E. Tavazzi, F.E. Munschauer, R. Zivadinov, Diffusion-weighted imaging predicts cognitive impairment in multiple sclerosis, Mult. Scler. 13 (2007) [30] E. Tavazzi, M.G. Dwyer, B. Weinstock-Guttman, J. Lema, S. Bastianello, R. Bergamaschi, V. Cosi, R.H. Benedict, F.E. Munschauer 3rd, R. Zivadinov, Quantitative diffusion weighted imaging measures in patients with multiple sclerosis, Neuroimage 36 (2007) [31] S.M. Smith, Fast robust automated brain extraction, Hum. Brain Mapp. 17 (2002) [32] C.C. Ford, Long-term experience with current disease-modifying drugs in multiple sclerosis, J. Neurol. 253 (Suppl. 6) (2006) vi37 vi44. [33] C.C. Ford, K.P. Johnson, R.P. Lisak, H.S. Panitch, G. Shifronis, J.S. Wolinsky, A prospective open-label study of glatiramer acetate: over a decade of continuous use in multiple sclerosis patients, Mult. Scler. 12 (2006) [34] M. Inglese, J.H. van Waesberghe, M. Rovaris, K. Beckmann, F. Barkhof, D. Hahn, L. Kappos, D.H. Miller, C. Polman, C. Pozzilli, A.J. Thompson, T.A. Yousry, K. Wagner, G. Comi, M. Filippi, The effect

8 68 R. Zivadinov et al. / Pathophysiology 18 (2011) of interferon beta-1b on quantities derived from MT MRI in secondary progressive MS, Neurology 60 (2003) [35] N.D. Richert, J.L. Ostuni, C.N. Bash, J.H. Duyn, H.F. McFarland, J.A. Frank, Serial whole-brain magnetization transfer imaging in patients with relapsing remitting multiple sclerosis at baseline and during treatment with interferon beta-1b, AJNR Am. J. Neuroradiol. 19 (1998) [36] M. Filippi, M. Rovaris, M.A. Rocca, M.P. Sormani, J.S. Wolinsky, G. Comi, Glatiramer acetate reduces the proportion of new MS lesions evolving into black holes, Neurology 57 (2001) [37] M.P. Sormani, M. Rovaris, P. Valsasina, J.S. Wolinsky, G. Comi, M. Filippi, Measurement error of two different techniques for brain atrophy assessment in multiple sclerosis, Neurology 62 (2004)