Volume 206, Issue 2 , Pages 199-202, 15 February 2003
Interferon beta 1a for secondary progressive multiple sclerosis☆
Article Outline
Abstract
This non-systematic review identified four randomised trials that have tested the efficacy of interferon beta in secondary progressive multiple sclerosis (SPMS). Two were trials of interferon beta 1a (IFNb1a) and two of interferon beta 1b (IFNb1b). All have shown significant reductions in relapse rates and accumulation of new magnetic resonance imaging (MRI) lesions, but only one trial (of IFNb1b) showed significant slowing of disability progression. Post hoc analyses of these trials suggest that the differences in outcomes might be partly explained by the differences between the trials in the proportions of patients with relapsing disease. In one of the trials of IFNb1a (the SPECTRIMS trial), the hazard ratio for progression in the treated relapsing patients with relapses in the two pre-study years was 0.74 compared to placebo patients with pre-study relapses and 1.01 in the treated patients compared to the placebo patients without pre-study relapses. In the same trial, the treatment effects on MRI parameters were more marked in the patients who had recent pre-study relapses compared with those who had not. These observations have led to the recommendation in national guidelines that prescribing of IFNb in SPMS be limited to those patients who have had disabling relapses in the last 2 years. These conclusions should be reviewed when the full results of all four trials have been published.
Keywords: Interferon beta, Secondary progressive multiple sclerosis
1. Introduction
The demonstration that IFNb reduces relapse rate and accumulation of magnetic resonance imaging (MRI) lesions in relapsing remitting MS led to the expectation that it might be effective in secondary progressive multiple sclerosis (SPMS) [1], [2], [3]. This expectation was heightened when it was shown in some trials in relapsing remitting MS that progression of disability was slowed [4], [5]. A Cochrane systematic review undertook a meta-analysis of the three pivotal trials and concluded that the relative risk of progression in relapsing remitting MS 2 years after randomisation to IFNb was significantly reduced, 0.69 (95% CI 0.55 to 0.87). However, the significance was lost under the worst (and unreasonably harsh) case scenario that IFNb-treated patients who dropped out were deemed to have progressed [6]. This review summarises the published results of the one trial of interferon beta 1a (IFNb1a) that has been published.
2. Methods
I searched PUBMED in December 2001 with “multiple sclerosis” and “interferon beta” and “randomised controlled trial” as the search terms. I also searched Issue 1 2002 of the Cochrane Library CENTRAL register with “multiple sclerosis” and “interferon” as the search terms limited by the year 2001. I discovered references to two randomised trials in SPMS, one of interferon beta 1b (IFNb1b) [7] and one of IFNb1a [8]. I did not discover the full-published results of two other trials, one of IFN1b [9] and one of IFNb1a (Goodman, 2001, personal communication), whose results have been presented at meetings.
3. Results
The SPECTRIMS trial was the only relevant published trial available for this review [8]. It was a parallel group double blind multicentre trial involving 618 participants. To qualify for entry, patients had to be aged 18–55, have clinically definite SPMS, an Expanded Disability Status Score (EDSS) between 3.0 and 6.5, and have had progressive deterioration during the previous 6 months. Treatment schedules were placebo or IFNb1a (Rebif) subcutaneously 22 or 44 μg three times weekly for 3 years. The groups were reasonably balanced for important demographic and disease-related variables by the randomisation process. About 92.4% of patients were followed for the planned 3 years, and 82% completed the 3 years of treatment.
The primary outcome measure was the time to progression by at least one point on the EDSS (or half a point if the EDSS was 5.5 or greater) confirmed 3 months later. There was no significant difference between the treated and placebo groups in the primary outcome measure. There was, however, an unexpected and unexplained interaction between treatment and sex such that the hazard ratio for female patients suggested highly significant benefit (hazard ratio 0.63 with 95% CIs 0.45–0.87), while for male patients it showed a trend towards an adverse effect (hazard ratio 1.30 with 95% CIs 0.85–2.01). In a post hoc analysis, there was a delay in the progression of disability in the patients who had experienced relapses during the 2 years before the study (Fig. 1, hazard ratio 0.74 compared with placebo, p=0.06) that was not seen in the patients without pre-study relapses (hazard ratio 1.01 compared with placebo, p=0.93). It should be emphasised that this analysis was not planned when the study was designed and both treatment doses were combined in this calculation.
Fig. 1.
SPECTRIMS trial. Kaplan Meier curves for time to confirmed EDSS progression for patients with and without relapses during the 2 years before the study.
The relapse-related secondary outcome measures showed highly significant results in favour of the IFNb1a-treated groups for both men and women. For instance, the mean (95% CI) relapse rate per year over 3 years was reduced from 0.71 (0.65–0.78) in the placebo group to 0.50 (0.44–0.56) in the 22-μg group and 0.50 (0.56–0.85) in the 44-μg group. This reduction was even greater in the patients who had relapses during the 2 years preceding randomisation.
The MRI-related outcome measures also showed highly significant results in favour of the IFNb1a-treated groups [10]. The burden of disease was measured as the median of the total volume of lesions on the T2 weighted scans. It increased by 10% during the 3 years of the trial in the placebo group but decreased by 0.5% in the 22-μg group and 1.3% in the 44-μg group (p<0.0001 for both doses). The median number of new or enlarging lesions on semi-annual T2 weighted scans was reduced from 0.67 per scan in the placebo to 0.20 in the 22-μg group and 0.17 in the 44-μg group (p<0.0001 for both doses compared to placebo). These effects were more marked in women than men and in those with pre-study relapses than those without compared to placebo.
Two hundred and eighty three patients had monthly gadolinium-enhanced scans during the first 9 months of the trial. In this cohort, there was a significant (p<0.005) reduction in both treated groups of new lesions detected by either gadolinium enhancement or appearance or enlargement of T2 weighted lesions. The median number of new lesions was 1.0 per scan in the placebo group 0.22 in the 22-μg group and 0.11 in the 44-μg group. In this analysis, there was a significant dose effect with a greater reduction in the 44-μg group than the 22-μg group (p=0.009). There was no difference in the effect for males and females for this outcome measure.
The differences in treatment effects on the MRI outcomes between the patients who experienced relapses in the two pre-study years and those who did not showed trends towards greater effects in those who had had relapses but were not statistically significant.
Neutralising antibodies to IFNb developed in 20.6% of the 22-μg group but less, as in the PRISMS study, of the 44-μg group (14.7%). Patients who developed antibodies showed a trend towards less reduction of relapse rate and had significantly less reduction in MRI burden of disease and rate of new lesions.
The adverse events in this trial were similar to those in other IFNb trials. Minor injection site inflammation was more common with IFNb1a than with placebo. Injection site necrosis occurred in 7 of 209 patients receiving 22 μg and 18 of 204 patients receiving 44 μg: it required surgical treatment in one case and led to treatment being stopped in three cases. Mild or moderate liver function abnormalities occurred more commonly with IFNb1a than with placebo.
4. Discussion
In the absence of published reports of another completed trial of IFNb1a (Avonex) in SPMS, it is not possible to compare the results in detail or combine results in a meta-analysis. Results presented at meetings and kindly made available by the authors (Goodman, 2001, personal communication) suggest that, as in the SPECTRIMS trial, there was no effect on disease progression measured with the EDSS. However, there was a significant effect on relapse and MRI-related outcome measures and on a new composite outcome measure combining walking speed, nine hole peg board completion speed, and cognitive function (paced auditory serial addition test). Therefore, despite evidence of a biological effect, especially convincing for the objective MRI outcome measures, the evidence suggests that IFNb1a has little, if any, effect on the progression of disability in SPMS.
The European IFNb1b trial which has been discussed at this symposium and extensively published revealed a significant effect not only on relapse and MRI-related parameters, but also on disability progression [7], [11], [12], [13]. At first glance, this difference in treatment effect from IFNb1a might argue in favour of a difference in efficacy or dose effect between the two forms of interferon. However, the American IFNb1b trial, which has not yet been published, apparently obtained results similar to those of the SPECTRIMS trial with significant effects on MRI and relapse-related outcomes but not on disability progression [9]. The negative result for disability-related outcomes in the American trial makes it more likely that the difference in results between the European IFNb1b trial and the other three trials was due to differences between the patient population in the European IFNb1b trial compared with those in the American IFNb1b trial and either of the IFNb1a trials. Relevant to this argument is that only 48% of the patients in the SPECTRIMS study had had relapses during the two pre-study years compared with 70% in the European IFNb1b study. In the SPECTRIMS trial, the patients who had pre-study relapses showed somewhat more marked benefits in terms of disability progression and a significantly greater reduction in relapses. The differences in the MRI outcomes between the patients who experienced relapses in the two pre-study years and those who did not were not significant.
The results of the trials available so far have led one national neurological society to recommend IFNb for SPMS if patients are not chair-bound (EDSS 6.5 or less) and have had two disabling relapses and minimal progression in disability due to slow progression in the last 2 years [14]. The American Academy of Neurology practice parameter recommends considering IFNb in SPMS if the patient is still experiencing relapses, but regards its effectiveness in patients with SPMS without relapses as uncertain [15]. Further conclusions must await full publication of the results of the two unpublished trials and meta-analysis of the results of all four trials. This would be more powerful if it were done by pooling individual patient data, which the investigators are exhorted to do. It is difficult to escape the sad conclusion that agents such as IFNb which reduce relapses and inflammation measured on MRI do not have significant effects on the process responsible for progression in MS. In short, the search for new more effective agents or combinations of agents must continue.
Acknowledgements
The author was chairman of the investigators' liaison committee and writing committee for the SPECTRIMS trial. He has received honoraria for lecturing and travel expenses to attend meetings from Biogen, Schering, Serono, and Teva. His department has received research grants from Serono.
References
- . Interferon beta-1b is effective in relapsing–remitting multiple sclerosis: I. Clinical results of a multicenter, randomized, double-blind, placebo-controlled trial. Neurology. 1993;43:655–661
- Intramuscular interferon beta-1α for disease progression in relapsing multiple sclerosis. Ann. Neurol. 1996;39:285–294
- . Randomised, double-blind, placebo-controlled study of interferon-beta 1a in relapsing–remitting multiple sclerosis. Lancet. 1998;353:1498–1504
- Impact of interferon beta-1a on neurologic disability in relapsing multiple sclerosis. Neurology. 1997;49:358–363
- . Long-term efficacy of interferon-beta-1a in relapsing MS. Neurology. 2001;56:1628–1636
- Interferon in relapsing–remitting multiple sclerosis (Cochrane Review). Cochrane Database Syst. Rev. 2001;4:
- . Placebo-controlled multicentre randomised trial of interferon β-1b in treatment of secondary progressive multiple sclerosis. Lancet. 1998;352:1491–1503
- . Randomized controlled trial of interferon beta-1a in secondary progressive MS. Clinical Results. Neurology. 2001;56:1496–1504
- . Interferon beta-1b in secondary progressive MS: clinical and MRI results of a 3-year randomized controlled trial. Neurology. 2000;54:2352
- . Randomized controlled trial of interferon beta-1a in secondary progressive MS. MRI results. Neurology. 2001;56:1505–1513
- . Effect of interferon-beta1b on magnetic resonance imaging outcomes in secondary progressive multiple sclerosis: results of a European multicenter, randomized, double-blind, placebo-controlled trial. European Study Group on Interferon-beta1b in secondary progressive multiple sclerosis. Ann. Neurol. 1999;46:850–859
- T(1) hypointense lesions in secondary progressive multiple sclerosis: effect of interferon beta-1b treatment. Brain. 2001;124(Pt. 7):1396–1402
- . Final analysis of the European multicenter trial on IFNbeta-1b in secondary-progressive MS. Neurology. 2001;57:1969–1975
- Association of British Neurologists, Guidelines for the use of beta interferons and glatiramer acetate in multiple sclerosis. Association of British Neurologists Report, 1–11 (2001).
- Disease modifying therapies in multiple sclerosis: subcommittee of the American Academy of Neurology and the MS Council for Clinical Practice Guidelines. Neurology. 2002;58:169–178
☆ Paper presented at European Charcot Meeting Venice October 2001.
PII: S0022-510X(02)00334-9
© 2002 Elsevier Science B.V. All rights reserved.
Volume 206, Issue 2 , Pages 199-202, 15 February 2003
