The results of the REGARD* study were presented at the recent 23rd Congress of the European Committee for Treatment and Research in Multiple Sclerosis (ECTRIMS). The REGARD study compared Rebif (interferon beta-1a) 44 mcg and Copaxone (glatiramer acetate) over 96 weeks in patients with relapsing-remitting MS. Principal investigator, Daniel Mikol, MD, PhD, from the Department of Neurology at the University of Michigan, Ann Arbor, provides some first-hand insight into the findings of the study.

• What was the overall rationale for the REGARD study and what information were you hoping it would provide?
  
• What were the entry criteria for patients enrolled in the REGARD study?
  
• Could you describe the overall study design?
  
• Were there any differences between the patients in the 2 arms of the study?
  
• What was the primary endpoint of the REGARD study, and what was the rationale for selecting that endpoint?
  
• With respect to the primary endpoint of time to first relapse, what did you see in the REGARD study?
  
• Why do you think there were fewer relapses overall than expected?
  
• What did the REGARD study show with respect to the safety of these medications?
  
• How do you feel the REGARD study advances what is known today about optimal treatment approaches for relapsing-remitting MS?

Q. What was the overall rationale for the REGARD study and what information were you hoping it would provide?

Dr Mikol: The REGARD study was the first well-controlled, randomized, head-to-head study between any interferon-beta and glatiramer acetate. The interferon-beta preparation used in the study was Rebif, and the study was sponsored by EMD Serono and Pfizer, co-marketers of Rebif in the US.

It's well accepted that the best way to gauge the relative efficacies of 2 agents in a disease such as relapsing-remitting MS is to compare them head-to-head. Prior to this study, the only way to try to compare interferon-beta and glatiramer acetate was by using results from the individual placebo-controlled studies — different trials with different patient populations, and trial designs that differed somewhat from one study to another. So in conducting the REGARD study, the intent was to use the same population of patients and the same study design to determine whether one agent was superior to the other.

Back to top

Q: You mentioned the importance of having the same population for both therapies — what exactly were the entry criteria for patients enrolled in the REGARD study?

Dr Mikol: The main inclusion criterion was relapsing-remitting MS. The study was planned in 2003 and began enrollment in 2004, so we used the 2001 McDonald criteria¹ for MS diagnosis which distinguishes it from previous trials that used the Poser criteria. Patients had to be between 18 and 60 years of age, had to have had at least one attack during the previous 12 months, and had to have a baseline disability rating (EDSS^†) between 0 and 5.5. Patients could not have progressive MS, and they could not have used any interferon-beta or glatiramer acetate prior to study entry.

Back to top

Q: Could you describe the overall study design?

Dr Mikol: The REGARD study was a multicenter, 96-week, open-label study. Patients were randomized to Rebif 44 mcg or Copaxone. Patients knew which of the 2 agents they received. However, evaluation of the results was blinded, both by MRI and by the neurological assessor — the neurological assessor rating EDSS, either at scheduled visits or unscheduled visits at the time of possible relapse, did not know which treatment the patient had been assigned to receive. MRIs were also analyzed in a blinded fashion.

A total of 81 clinical centers in 14 different countries took part in the study. The initial design called for 736 patients. This number of patients was expected to provide 80% power to detect a 30% relative difference in the primary outcome measure. With this subject size, it was projected that approximately 460 patients would have at least one relapse during the course of the study; 764 subjects were actually enrolled (386 Rebif 44 mcg; 378 Copaxone).

• An open-label, randomized, multicenter, comparative, assessor-blinded study
• Planned sample size of 736 patients provides 80% power to detect 30% relative increase in primary parameter (time to first relapse at 96 weeks)
• Assumes 460 patients will have at least one relapse over 96 weeks

Patients were seen at baseline and randomized to receive either Rebif 44 mcg tiw SC or Copaxone 20 mg qd SC. The main clinic visits for neurological assessment were conducted at 6-month intervals, when patients underwent EDSS assessment. All patients had a baseline brain MRI, and approximately 60% of patients underwent serial MRIs every 6 months.

Back to top

Q: Were there any differences between the patients in the 2 arms of the study?

Dr Mikol: For the most part, there were no differences at baseline — in both groups, patients were about 37 years of age, and roughly 70% of patients in each group were female. The mean EDSS at entry was 2.3 in both groups, which is similar to other recent relapsing-remitting MS trials. Gadolinium-enhancing volume and T2 lesion volume were not significantly different between groups.

A statistically significant difference was that more patients in the Copaxone group (48%) had received steroid treatment in the 6 months prior to the study than patients in the Rebif group (40%; P=0.023). Another difference that was observed regards the proportion of patients who had experienced more than one relapse in the 24 months prior to enrollment. While not statistically significant it is important to note that 71% of Copaxone-treated patients had experienced more than one relapse, compared with 60% of Rebif-treated patients (P=0.057). It is difficult to extrapolate how this difference might have affected outcome, if at all.

Back to top

Q: What was the primary endpoint of the REGARD study, and what was the rationale for selecting that endpoint?

Dr Mikol: The primary endpoint in the REGARD study was time to first relapse during the 96-week treatment period. Many other relapsing-remitting MS trials have used relapse rate, but those trials have typically used a placebo comparator. This endpoint was selected because it was thought to be clinically meaningful. I believe that the type of patient entering this study (ie, just beginning treatment) has concerns about their future disease course, such that delaying the time to relapse is very important.

Back to top

Q: So with respect to the primary endpoint of time to first relapse, what did you see in the REGARD study?

Dr Mikol: In terms of the primary endpoint — time to first relapse — there was no significant difference between the 2 agents over the course of 96 weeks. The hazard ratio, which is a measure of the relative chance of experiencing a relapse on one agent versus the other, was just very slightly in favor of Rebif (0.943). But the confidence interval was very broad, and there was certainly no statistically significant difference between the 2 agents (95% CI, 0.74-1.21; P=0.643).

• Time to first relapse: hazard ratio (95% CI) = 0.943 (0.74, 1.21)
• This was not statistically significant (P=0.643), so the study did not meet its primary endpoint
• 460/736 patients (63%) were expected to have ≥1 relapse
• However, only 258/764 patients (34%) experienced ≥1 relapse
  • Approximately 45% fewer relapses than expected (258/460)

What was unexpected in this study was the small number of relapses overall. Based on previous studies, it was expected that around 63% of patients would have had at least one relapse over the course of 96 weeks, and in fact only 34% had any relapse — a little over half of the number that were expected. As a result, the study was not powered sufficiently to show a potential difference between the 2 agents with respect to this outcome measure.

Back to top

Q: Why do you think there were fewer relapses overall than the 63% that you expected?

Dr Mikol: Well, that is an interesting and important question, and we don't have a definitive answer. If you follow the trend in clinical studies with interferon-beta or glatiramer acetate over recent years, the relapse rate seems to be declining over time, with other recent studies yielding lower relapse rates than have been seen historically. In addition, there are a number of possible reasons why we saw fewer relapses in this study compared to earlier trials, which reflects study design. Whereas the REGARD study required just one relapse in the previous 12 months, many previous trials required subjects to have had at least 2 relapses in the previous 2 years. So we may have biased towards a less active group of patients in REGARD.

Moreover, we now have a number of available treatment options for relapsing-remitting MS, as well as an increasing number of clinical trials. Patients in the REGARD study had to be naïve to treatment with interferon-beta or glatiramer acetate. In earlier trials, these treatments were not as widely available as they are now, especially in some parts of the world, so at that time treatment-naïve patients may have had very active disease. Today, however, you would expect the majority of patients with active disease to already be on one of these treatments. So it is quite probable that we were trying to detect a treatment benefit in a population that had a relatively low level of disease activity. The drop-out rate was also higher than expected, which means that complete data were available from fewer patients than we had originally projected.

Back to top

Q: What did the REGARD study show with respect to the safety of these medications?

Dr Mikol: Neither drug appeared to offer any safety advantage over the other. There were no unexpected adverse events, and both agents were rather comparable in terms of the incidence of any treatment-related adverse events. The overall withdrawal rate from the study was higher than historical trials — 17% (21% on Rebif and 14% on Copaxone). Many of the incremental drop-outs in the Rebif group were patients who were lost to follow-up or whose reason for withdrawing from the study was described as "other," rather than any specific drug-related effect. In the Rebif group, 23 patients (6%) discontinued treatment due to adverse events, and in the Copaxone group, 18 patients (5%) discontinued treatment due to adverse events.

Around 95% of adverse events in both groups were mild to moderate, meaning that about 5% in each group were more severe.

Certain adverse events were more commonly seen in the Rebif group, and others were more commonly seen in the Copaxone group. In the Rebif group, not surprisingly, flu-like side effects, headache, and myalgias were more common, as well as elevated ALT levels. Patients in the Copaxone treatment arm more commonly experienced injection-site pruritis, injection-site swelling, and injection-site induration, as well as dyspnea and postinjection systemic reactions. 19 Copaxone patients (5.1%) reported postinjection reactions vs 0 Rebif patients. Otherwise, there were no statistically significant differences with respect to adverse events between the 2 treatment arms, including depression, which was not significantly different.

Back to top

Q: Finally, then, how do you feel the REGARD study advances what is known today about optimal treatment approaches for relapsing-remitting MS?

Dr Mikol: This study is important because it is the first large, randomized, controlled, head-to-head comparison between any interferon-beta and Copaxone in relapsing-remitting MS. As far as advancing knowledge about the relative clinical efficacy of Rebif and Copaxone, I don't think that we can draw any definitive conclusions with respect to the primary endpoint in this study. With far fewer observed relapses than expected overall, the study was not sufficiently powered to see the projected difference in time to first relapse for one treatment arm versus the other. I think that investigators in other studies will have to take into account the possibility that treatment-naïve patients who are available for enrollment in clinical trials today may not have as active disease as patients who were enrolled in previous studies, conducted at a time when effective treatment options were not widely available.

Ultimately, in order to show a treatment effect on a clinical outcome (ie, relapses) in a clinical trial, there must be a sufficient number of those events. If the McDonald criteria are employed, as they were in this study, patients who have had just one clinical attack (and demonstrated MRI change) can be enrolled. However, it is not necessarily possible to predict how such patients are likely to behave during the course of the study: Patients who have had multiple attacks, but only one in the past year, may not have another attack during the course of the study. As a result, there may need to be some rethinking about what criteria are most appropriate when enrolling subjects into a 2-year clinical study.

Medscape