Effects of glatiramer acetate on relapse rate and accumulated disability in multiple sclerosis: meta-analysis of three double-blind, randomized, placebo-controlled clinical trials

Abstract

Three randomized, double-blind, placebo-controlled trials have shown that glatiramer acetate (GA) is effective in reducing relapse rate in patients with relapsing-remitting (RR) multiple sclerosis (MS). Using raw data pooled from 540 patients, we performed a meta-analysis of these three trials, to investigate whether the extent of G A efficacy varies according to disease-related variables at study entry. Three regression models were developed to assess the efficacy of G A on the annualized relapse rate (primary outcome measure), on the total number of on-trial relapses and on the time to first relapse. We also explored the efficacy of G A on accumulated disability and the potential role of baseline clinical variables as predicto rs of relapse-rate variables and treatment efficacy. The mean adjusted annualized relapse rate on study was 1.14 in the pooled placebo -treated subjects and 0.82 in the pooled GA group (P =0.004), indicating an average reductio n in annualized relapse rate of 28%. A bout a one third reductio n of the total number of on-trial relapses was also observed in patients receiving GA (P B-0.0001), who had a median time to the first relapse of 322 days versus a median time to the first relapse of 219 days seen in those receiving placebo (P =0.01). A beneficial effect on accumulated disability was also found (risk ratio of 0.6; 95%; C I =0.4-0.9; P =0.02). The drug assignment (P =0.004), baseline EDSS score (P =0.02) and number of relapses during the two years prior to study entry (P =0.002) were significant predicto rs of on-trial annualized relapse rate. No other demographic or clinical variable at baseline significantly influenced the treatment effect. This meta-analysis reaffirms the effectiveness of G A in reducing relapse rate and disability accumulatio n in RRMS, at a magnitude comparable to that of other available immunomodulating treatments. It also suggests that G A efficacy is not significantly influenced by the patients’ clinical characteristics at the time of treatment initiation.