Reports

This report is based on medical evidence presented at sanctioned medical congress, from peer reviewed literature or opinion provided by a qualified healthcare practitioner. The consumption of the information contained within this report is intended for qualified Canadian healthcare practitioners only.

23rd Congress of the European Committee for Treatment and Research in Multiple Sclerosis (ECTRIMS)

Prague, Czech Republic / October 11-14, 2007

Thanks to several new head-to-head studies which compare the two first-line treatment alternatives, the speculation about the relative efficacy and tolerability of glatiramer acetate (GA) and interferon beta-1 (IFNß-1) as first-line immunomodulatory agents in the treatment of relapsing-remitting multiple sclerosis (RRMS) seems to be coming to an end. New evidence suggests that GA is as potent in the short-term as either the current subcutaneous (s.c.) formulation of IFNß-1a or IFNß-1b when given at the usual doses.

Outcomes of the REGARD Study

The REGARD (Rebif vs. Glatiramer Acetate in Relapsing MS Disease) study was a 96-week, randomized, multicentre, multinational, open-label, trial comparing the efficacy and safety of IFNß-1a s.c. 44 µg three times weekly and GA 20 mg s.c. daily in a total of 764 patients with RRMS diagnosed according to the McDonald criteria. To be eligible for inclusion, patients had to have experienced at least one attack in the previous 12 months and be clinically stable. Neurologists who evaluated the patients’ Expanded Disability Status Scale (EDSS) were blinded to the treatment. All patients had baseline magnetic resonance imaging (MRI) and a subset of 416 patients (about 60%) underwent MRI every six months. Baseline EDSS (mean, 2.34) and other characteristics were similar between the two groups; the only notable difference was that 71% of patients randomized to receive GA had had a relapse in the previous 24 months compared with 60% of the those assigned to IFNß-1a treatment.

Time to first relapse, the primary outcome measure, did not differ significantly between groups: 495 and 432 days for the IFNß-1a and GA arms, respectively (hazard ratio [HR] 0.943; P=0.643). “The Kaplan-Meier curves are very similar throughout the 96-week duration of the study for both treatment groups,” observed Dr. Daniel D. Mikol, Director, Multiple Sclerosis Clinic, and Assistant Professor, Department of Neurology, University of Michigan, Ann Arbor, US, who presented the findings on behalf of the REGARD study group (Figure 1).

Figure 1. REGARD: Time to First Relapse

Similarly, for the main secondary end point, the number of active T2 lesions per patient per scan did not differ significantly between treatment groups: 0.7 and 0.8 for IFNß-1a and GA, respectively (P=0.178). The tertiary end point, annualized relapse rate, was also similar between groups (0.3 and 0.29 for IFNß-1a and GA, respectively, P=0.828). In the overall population and the MRI subset, the only differences to reach statistical significance were in the number of T1-weighted gadolinium (Gd)-enhancing lesions (0.2 and 0.4, P<0.001).

Prespecified analysis of six subgroups revealed significant differences in favour of IFNß-1a in those patients with baseline EDSS below the median (HR 0.648; P=0.022). However, these subgroup results were criticized during the subsequent question-and-answer session. First, it was pointed out that the subgroup data had not been corrected for multiplicity; the term statistically significant was therefore inapplicable to the differences observed. Second, since the patients with EDSS below 2 had appeared to perform better on IFNß-1a, the more severely disabled patients may benefit more on GA. Third, although the number of on-study relapses was lower than anticipated, the existing Kaplan-Meier curves were superimposable and hence would have been unlikely to diverge even with more events or over a longer follow-up, reinforcing the similarity of efficacy for the two treatments.

There were no unexpected adverse events during the REGARD study. However, of the 374 patients receiving IFNß-1a who had antibody data available at their last assessment at 96 weeks, 102 (27.3%) were neutralizing-antibody (NAb)-positive and 125 (33.7%) had a positive NAb test at least once during the study. Significantly more patients receiving IFNß-1a than GA experienced flu-like illness (P<0.001), headache (P<0.001), myalgia (P=0.014) and alanine aminotransferase abnormalities (P=0.002); with GA, only pruritus, swelling, induration and reactions at the injection site, and dyspnea were significantly more frequent than with IFNß-1a (P<0.001, P<0.001, P=0.005 and P<0.001, respectively). Overall, 13.6% of patients receiving GA failed to complete 96 weeks of treatment compared with 20.9% of those using IFNß-1a.

BECOME Study Results

Final findings from the BECOME (Betaseron vs. Copaxone in MS with Triple-Dose Gadolinium and 3-T MRI Endpoints) study were also presented here at ECTRIMS. BECOME is the first comparative, randomized study to focus on MRI outcomes of patients treated with GA and IFNß, in this case IFNß-1b.

BECOME randomized 75 patients with MS to one of two treatment arms: IFNß-1b 250 µg s.c. every other day or GA 20 mg s.c./day. Monthly MRI scans and other measures were used to monitor treatment efficacy for up to 24 months. All MRI scans were conducted by the same 3-T-dedicated unit according to the same protocol and read in a blinded fashion by the same experienced neuroradiologist, who counted the number of enhancing lesions and new T2 lesions to determine the mean number of combined active lesions (CALs) per scan, the primary outcome measure. The mean number of CALs per month and number of new enhancing lesions (NELs) on a given scan were a secondary outcome measure. Additional outcome measures were: the change in neurological function, as measured by the EDSS and the MS Functional Composite (MSFC) every three to four months; change in cognitive function as measured by formal neurocognitive testing at month 0, 6, 12, and 24; and the development of relapses. All clinical assessments were conducted by neurologists masked as to treatment assignment. Both treatment groups were similar at baseline.

Post-treatment lesion counts for months 1 to 24 showed no statistically significant differences between the groups although the results consistently favoured the GA-treated group. The median CALs per scan was 0.62 for the GA group vs. 0.78 for IFNß-1b-treated patients. Mean CAL values (standard deviation [SD]) were 1.75 (2.76) and 2.67 (5.14), respectively. In the GA group, nine patients (23%) remained lesion-free compared with six (17%) in the group receiving IFNß-1b. Likewise, median CALs per month did not differ significantly between groups, although again, the GA-treated group fared somewhat better with a median count of 0.38 vs. 0.60. Mean (SD) values were 1.27 (2.32) and 1.67 (2.84) in the GA and IFNß-1b groups, respectively. The median number of NELs was 0.75 for the GA patients and 1.19 for those who had received IFNß-1b; mean (SD) values were 0.84 (1.64) and 1.09 (1.80), respectively; 10 (26%) GA patients and seven (19%) IFNß-1b-treated patients remained NEL-free. Although the monthly CAL counts showed a significant decrease on starting therapy over the first 12 months in the IFNß-1b group but not in the GA group, this was the only significant difference between the two treatments. Moreover, consistent with these results, none of the clinical outcome measures (relapses, EDSS, MSFC) differed between groups following treatment. Both groups experienced a marked drop in annualized relapse rates from a median of approximately 2 to around 0.3 and only a small and similar proportion (<10%) developed sustained progression in disability as measured by the EDSS. Both treatment groups significantly improved in global scores, the nine-hole peg test and the three-second Paced Auditory Serial Attention Test (PASAT) on the MSFC. There were no significant differences between groups in cognitive function changes over time. The investigators concluded, “The superiority of IFNß over GA in reducing the incidence of acute inflammation in MS may have been overestimated.”

Effect of MS on Brain Volume

Brain atrophy, as explained by Prof. David Miller, Institute of Neurology, London, UK, “is likely to reflect neuroaxonal loss; axons contribute about 50% of the bulk of white matter, neurons contribute the majority of the bulk of grey matter.” In MS, he pointed out, “atrophy has been seen at all stages from CIS [clinically isolated syndrome] to advanced progressive MS; on average about 0.5 to 1% loss of brain volume per year.” But he cautioned delegates: “One needs to bear in mind other factors including the effect of inflammation to increase brain volume and anti-inflammatory therapy which may reduce brain volume—at least in the short term.” MRI-based methods such as the SIENA (Structural Image Evaluation, using Normalization, of Atrophy) registration technique, which is applicable to small sample sizes and the cross-sectional version of SIENAX, a global segmentation technique which isolates the brain from surrounding structure, are particularly useful, he added.

A team of investigators led by Dr. Omar Khan, Professor of Neurology, and Director, MS Clinical Research Center, Wayne State University School of Medicine, Detroit, Michigan, US, has used both these methods to examine the long-term effect of GA, IFNß-1b s.c., and IFNß-1a s.c. and i.m. in early and minimally disabled patients with RRMS. The use of long-term data, as indicated by Prof. Miller, helps avoid the possible confounding effects on apparent brain volume of edema resolution in the first year of therapy. Each of the 309 patients had MRI scans at baseline, and after five years’ continuous therapy, on the same scanner, the same neurologist recorded their EDSS at the same time points. “Among the three therapies compared in our study, GA therapy had the best effect in reducing the rate of brain atrophy over five years, which was significantly better than IFNß i.m. or IFNß s.c.,” the investigators concluded (Figure 2).

Figure 2. Effect of Therapies on
ophy Over Five Years

<img1081|center>

An additional study independently confirms previous research indicating that GA but not IFNß inhibits the evolution of T1-weighted lesions into black holes, considered as markers of irreversible tissue loss. Dr. Nancy Richert, National Institutes of Health, Bethesda, Maryland, US, and colleagues used magnetization transfer imaging (MTR) to follow the evolution of contrast-enhancing lesions (CELs) into black holes in 21 RRMS patients during a three-month baseline period followed by a 12-month treatment period with GA. They found that whereas the proportion of CEL-derived black holes increased in the IFNß group, from 21.2% to 26.0%, it significantly decreased with GA treatment, from 20.8% to 4.8%.

These findings confirm those obtained previously with GA in a large, placebo-controlled trial conducted by Dr. Massimo Filippi, Director, Neuroimaging Research Unit, Scientific Institute and University, Ospedale San Raffaele, Milan, Italy, and colleagues.

Long-term Follow-up of MS Patients

These results on brain tissue integrity suggest that GA may have an in situ effect independent of the blood-brain barrier, possibly leading to prevention of irreversible central nervous system (CNS) tissue injury. This, in turn, may slow or prevent the long-term progression of disability, as indicated by two ten-year follow-up studies of this agent. In one of these, an ongoing, prospective study, ten years of continuous GA therapy showed suppression of relapse rate to approximately one relapse every five years, with 62% of patients showing stable or improved disability levels. Although average disease duration was 17 years, 92% were able to walk without help. “It was… very reassuring to hear that those patients who were taking therapy for this many years did actually quite well,” said Dr. Khan, who reviewed the results.

The other study, conducted by a group led by Dr. Adriana Carrá, Hospital Británico de Buenos Aires, Argentina, and colleagues, compared two patient cohorts. In one group, 174 patients had received at least one dose of GA since inclusion, which comprised 112 patients (64.3%) remaining in the study and 62 patients (35.6%) who withdrew from the study before March 2007 having received at least one dose of GA. Of the 62 who withdrew, 32 (52.5%) returned for the long-term follow-up visit. The second natural-history cohort comprised an untreated control group of 508 patients who voluntarily refused, or were unable to receive, treatment. Annualized relapse rate was reduced significantly, by 86%, compared with baseline (P<0.0001) for all GA-treated patients to approximately one relapse every 4.7 years; 56.9% of patients remained relapse-free. Nearly three-quarters (73%) of the continuously GA-treated patients showed either no change in, or had improved, their EDSS score and 76% remained ambulatory after a mean of 6.7 ±2.6 years. Progression to EDSS scores of 4.0 and 6.0 was significantly less for the patients on continuing GA vs. the natural-history cohort (P<0.0001). This finding suggests that “GA treatment modifies disease progression [in the long term],” the researchers concluded.