Promises, promises…

But I have promises to keep, And miles to go before I sleep, And miles to go before I sleep. —Robert Frost, From “Stopping by Woods on a Snowy Evening” Patients with primary progressive multiple sclerosis (PPMS) are distinguished clinically by the absence of the key hallmarks of multiple sclerosis (MS): relapses and remissions. This, together with a tendency toward more localized involvement of the central nervous system and a predilection for the spinal cord, resulting in a slowly progressive paraparesis (and occasionally a hemiparesis), tends to characterize this subset of patients with MS.1 The separation of PPMS from other forms of the condition evolved further in the early 1980s, primarily as a result of imaging studies that were seen as useful contributions to our understanding of the complex process that is MS.2 Consistent findings indicate that patients with PPMS have fewer lesions on magnetic resonance imaging (MRI) of the brain and, perhaps more importantly, less enhancement after injection of the contrast agent gadolinium-DTPA. Subsequent studies have suggested that there may be more spinal cord involvement, with perhaps a greater incidence of diffuse change in the cord.3, 4 The suggestion from MRI that PPMS may be a less inflammatory condition has been supported, to some extent, by pathological studies that have characterized four distinct patterns of immunopathogenic damage.5, 6 The fourth of these, predominantly primary oligodendrocyte degeneration, appears to be more commonly seen in patients with PPMS, though other patterns are seen.7 Over this time an attempt was made to develop diagnostic criteria that would be sufficiently rigorous to ensure consistent recruitment to research studies and particularly clinical trials.8 These criteria required the presence of abnormalities in the cerebrospinal fluid. They were incorporated virtually unchanged into the McDonald criteria,9 but abnormalities in the cerebrospinal fluid are no longer essential in the revised criteria.10 Ironically, the improved characterization of PPMS has resulted in patients with this form of the condition being excluded from most therapeutic trials of MS. This is partly because they are considered by some to be a different form of MS (rather than part of a continuum), and more practically because the primary outcomes for such trials, such as the “time to” or frequency of relapses or the development of new lesions on MRI, enhancing or simply T2, over time, are either not relevant or poorly responsive in PPMS. Despite these restrictions, two pilot studies of disease-modifying treatments have been performed involving intramuscular interferon-β-1a and subcutaneous interferon-β-1b.11, 12 No safety issues were detected in these studies, but although completely underpowered to show effect, there was a suggestion of some difference between the placebo and treated arms on MRI and the MS Functional Composite. It is difficult to know how much weight to place on these observations, and to date, no phase III trial of either agent has been planned.13 It was thus a great boon that a phase III trial of glatiramer acetate was announced and subsequently planned.14 The rationale for the study appeared to be based as much on hope and opportunity as sound scientific evidence. There was a suggestion from a post hoc analysis of a small number of patients (31) with chronic progressive MS that there may be a delay in the time to progression of disability.15 The investigators powered their study on natural history data available at the time from London, Ontario, Canada,16 and attempted to maximize the likelihood of demonstrating an effect by ensuring that 40% of the enrolled patients would have an entry Extended Disability Status Scale (EDSS) score of 3.0 to 5.0 (stratum I), 50% of whom they predicted would progress by one level each year and 60% would have an EDSS of 5.5 to 6.5 (stratum II), 20% of whom were anticipated to progress by one level each year. Assuming glatiramer acetate would prolong the time to clinical progression by 40%, a 40% dropout rate, and a 2:1 randomization of active drug to placebo, the recruitment sample was estimated at 900. In fact, 943 patients were randomized, and they divided favorably between the 2 strata in that 54% were in stratum I and 46% were in stratum II. This should have augured well for the study, but for reasons that remain unclear, the progression in stratum I was more in keeping with a recent natural history study from British Columbia17; the percentage of patients that progressed was only 16%, rather than the 40% that had been anticipated. Perhaps this result serves to remind us that patients in trials often fare better than those in natural history studies. The study was terminated by the data safety monitoring committee after the second preplanned interim analysis, at which time 757 patients had completed at least 2 years or had terminated prematurely. The decision to stop any trial prematurely is never taken lightly. In a patient population where there is no effective treatment available, the data monitoring safety committee would need to have been confident that there was virtually no possibility of demonstrating benefit before deciding to bring the study to a close. Terminating a study early often results in controversy18 with the concern that longer duration would demonstrate benefit or, as in the case of the study of sulfasalazine in MS, benefit initially seen would disappear over time.19 In relation to the latter, the authors called for caution in accepting efficacy measured by changes in clinical parameters in trials with less than 3 years of follow-up. It appears highly unlikely that the results of this study would have been any different had it run its course. The primary end point, an intention-to-treat analysis of the time to confirmed disease progression, showed no difference between the two groups, and there was no difference seen in the secondary clinical end points. A suggestion from a post hoc analysis that benefit in the primary outcome may be seen in male patients was explained by the fact that male patients taking placebo did somewhat worse during the study. Differences in the MRI outcomes studied were also disappointing. A possible effect on the number of gadolinium-enhancing lesions seen in the first year disappeared gradually during the study. In short, the results reported in this issue of the Annals of Neurology only go to reinforce the view that such agents may have little role in this particular form of PPMS. They do, however, encourage consideration of other approaches. PPMS is an important model of progression20 and warrants intense study, but perhaps from a different perspective. If our understanding of the pathological processes underlying progression is correct, then urgent consideration of neuroprotective agents and repair strategies is warranted. These approaches, which are ideally suited to PPMS, have been slow to get started partly because of the complexities inherent to remyelination and partly because of the challenges of trial design, recently addressed in a National MS Society meeting, entitled “The Future of MS Therapies: Redesigning MS Clinical Trials in a New Therapeutic Era,” held in Washington, DC.21 In summary, as is so often the case, this trial has told us as much, if not more, about the study population than the agent under study. It has provided important insights into PPMS and has provided useful hints in the designing of future trials. It has also given a strong indication of the future direction of therapeutics for this patient population. One can only hope that the journey to reach an effective treatment is not as depressingly long as Robert Frost would suggest. Having been excluded from most clinical trials, it is vitally important that patients with PPMS have a realistic hope of new and effective treatments in the future.