Edited by Richard A. Rudick and Donald E. Goodkin. 574 pp. London, Martin Dunitz, 1999. £ 79.95. ISBN 1-85317-812-8
The New England Journal of Medicine -- May 18, 2000 -- Vol. 342, No. 20
It was not long ago that the combination of the words "therapeutics" and "multiple sclerosis" verged on the oxymoronic. If there are investigators who still have this point of view, they are not represented among the authors of the 40 chapters in this well-produced book. The many topics discussed include clinical research methods, outcome measures, and the fine details of magnetic resonance imaging and spectroscopy. It is a sign of the times that three chapters describe the process of drug approval in the United States, Canada, and Europe. The general stance of most of the authors is remarkably upbeat, perhaps reflecting the prolonged dearth of reproducible advances in the treatment of this often tragic disease.
There is much useful information in this book, and the enthusiasm is refreshing, even for students of multiple sclerosis whose ears ring with the cries of "wolf." However, it would be easy to gain the impression that treatment with a wide variety of agents has had a major effect on this most difficult disease. Only a single chapter, by Rice, reports negative results but does so clearly and definitively.
In the first chapter, the editors appropriately state, "No surrogate measure has acceptable validity for predicting eventual disability." However, there is a relative absence of reservations, caveats, critiques, and contradictions, which would have balanced the generally enthusiastic tone of the text. Readers will have to look elsewhere for thoughtful discussions of the limitations of existing data (all very short-term), the pitfalls of the surrogate markers on which great reliance is placed, and the potential methodologic limitations of the numerous studies with positive results.
There is no accounting for taste, but the index does not contain the entries "blinding," "dropouts," or "withdrawals." Some of the findings of research into this complex disease can be difficult to interpret by persons outside the field and even by those within it. But the opinions of investigators directly involved in specific studies, who are familiar with the soft underbelly of otherwise definitive data, often provide unique insights. Unexpectedly, the fact that the results of so many well-designed, randomized trials involving multiple sclerosis have not been reproducible is neither explained nor acknowledged.
A chapter on standardized reporting leaves readers wondering why the contributors have not, in some of their own studies, followed the useful advice they offer here. In "The Negative Clinical Trial," readers are told that the negative results of many trials may really be positive. "Several studies have illustrated the powerful placebo effect in [multiple sclerosis] patients, which makes it difficult to demonstrate a significant beneficial effect of the agent to be tested." This perspective is not supported by any data and is at odds with the rate of deterioration in placebo groups, which at least equals and in many positive trials surpasses rates in studies of the natural history of the disease. Included in the table of reasons for negative clinical-trials results are "Type 1 error" and "technical problems, including unblinding of examiners and patients." It is stated in the same chapter that copolymer 1 "was beneficial in one center and not in the other whereas the larger multicenter trial led to [Food and Drug Administration] approval." Perhaps someday type II errors will be shown to have been common in studies of multiple sclerosis, but readers are not told in this instance that the center-specific effect concerned one indication (the progression of disease), whereas approval for efficacy was for another (the prevention of relapse).
Suitable sobriety is maintained in the chapter by Confavreux on combination therapies. However, Frohman's enthusiasm for treating patients with relapsing and remitting multiple sclerosis is unbridled, and the author provides a detailed protocol for the use of cyclophosphamide, a treatment that has been unsuccessful in the only two blinded, controlled studies of its effects.
It is difficult to read this book without being concerned about the general dwindling of government funding for universities and for clinical trials. Is independent-mindedness among clinical investigators in jeopardy? The key words in this evolving story are striking: a tragic disease, expensive therapy, surrogate markers, and short-term efficacy. Even short-term efficacy -- an imperfect goal, to be sure -- will be difficult to demonstrate, because under most circumstances, new therapies must be validated in relation to existing ones. Although it is recognized that the use of existing therapies is supported by studies in which the period of evaluation was often less than two years and rarely as long as three, how much more difficult will it be to make advances in treatment? New drugs will need to be more effective and will pose greater risks with respect to capital expenditures by industry, in the context of a shortage of untreated patients.
The views expressed in this book reflect the high spirits accompanying
the emergence of the field from decades of skepticism, false dawns, and
considerable frustration. The hope is that the next few years will see
the accumulation of data on longer-term outcomes, so that the high spirits
in this book will prove to have been justified.
George C. Ebers, M.D.
Oxford OX2 6HE, United Kingdom