(Stroke. 1995;26:2216-2218.)
© 1995 American Heart Association, Inc.
Articles |
Acute Stroke Treatment Trials in the United States
Rethinking Strategies for Success
From the Department of Neurology, Medical College of Georgia, Augusta (R.J.A.); the Medical Center of Central Massachusetts, Worcester (M.F.); the Cleveland Clinic Foundation, Ohio (A.J.F.); and the Department of Molecular and Experimental Medicine, The Scripps Research Institute, San Diego, Calif (G. del Z.).
Correspondence to Robert J. Adams, MD, Department of Neurology, HB 2060, Medical College of Georgia, Augusta, GA 30912-3200. E-mail @emgmhs.mcg.edu.
 |
Introduction |
|---|
 Top Introduction References |
|---|
In the recent Humana Lecture,1 Harold Adams quoted what might be the prevailing sentiment guiding the search for a treatment for acute ischemic stroke: "When we have a treatment for stroke, we will know it." The implicit assumption is that there is a treatment that will dramatically improve acute stroke outcome. The current approach to finding a treatment for ischemic stroke is based on single-agent trials, sponsored mainly by pharmaceutical companies, that may produce a "winner," to be followed by a "therapeutic cocktail" to combat the complex events set in motion by cerebral ischemia. Progress in ischemic stroke treatment trials over the last two decades has included improved statistical design, radically shortened entry windows, standardized rating instruments, and shared sponsorship between industry and government. We wish to raise concerns about the current approach to testing new stroke treatments and the impact that changes in healthcare delivery in the United States may have on the conduct of acute stroke trials. The success of clinical research in the treatment of acute stroke depends not only on trial design but on local, regional, and national factors that determine the environment in which acute stroke care is provided and that affect the feasibility of trial execution. In this editorial, we present some of the important issues facing those who design and implement acute stroke treatment studies. Our opinions pertain primarily to the medical environment in the United States. Our goal is to stimulate discussion and suggest a forum for the critical evaluation of trial design issues and also the development of a national strategy for stroke.
We are concerned about the optimal use of critical resources needed to conduct stroke treatment trials. By "resources" we mean the investigators and stroke research teams; acute stroke patients available for enrollment; cooperative referring doctors and emergency-services personnel; data management; and sufficient funding for all the personnel, drug, training, travel, marketing, and educational efforts involved in these studies. The rate-limiting resource in most trials is the patient with a moderately severe deficit who arrives at the participating-site hospital within a few hours of symptom onset. Clinical trials typically are conducted at tertiary centers that attract more complicated patients who are more likely to be excluded for medical or other reasons. Some centers involve community hospitals in acute stroke trials. These collaborations, however, require great effort to establish and maintain. Referral to a tertiary center often involves delays that are incompatible with the short entry windows now used, and the cost of transfer may not be reimbursable. Financial incentives offered to the investigator for early entry may not increase early enrollment because much of the patient-entry delay depends on other factors such as patient response, emergency medical services, and referral patterns.2
Multiple concurrent studies compete for the same patients and often test similar classes of agents, reflecting competition among pharmaceutical sponsors more than rational application of preliminary data. Given the difficulty of supporting a dedicated research team, a center's decision to participate may be influenced more by financial considerations than the merits of the agent or trial design. In some cases, centers participate in as many as eight acute ischemic trials concurrently.4 While this "recent profusion of stroke treatment trials"4 is in some sense very encouraging, it raises the prospect that more cooperation and less competition might get us to the goal of effective therapies at a faster pace by concentrating the limited patient resource in a smaller number of trials. Community hospitals and academic centers should be encouraged to develop regional acute stroke referral networks that optimize care, minimize negative competitive pressures, and enhance trial participation. Such networks could be modeled on regional trauma systems, which attempt to ensure that the urgent care of the severely injured patient is determined not by chance, local politics, or market forces, but by regional plans that designate skill levels by use of standard criteria. This system has led to prompt care for many trauma victims and could be applied to the care of acute stroke patients as well.3
Restricted choice for health care and increasing pressure for early hospital discharge affect the feasibility of stroke trials. The policies of managed-care providers will have an increasingly important effect on trial participation, recruitment, and execution and have to be considered. More effective incentives may be needed for community physicians and hospitals to support and participate in stroke trials, particularly if the healthcare delivery environment becomes more hostile to clinical research. The National Institutes of Health has recently impaneled a group of experts to make recommendations regarding the effects of cost-conscious managed-care organizations on clinical research in general.5 The potential effect on stroke treatment trials should be specifically considered.
It could be argued that patient accrual will increase once a treatment is approved, to "get them in the door early." However, an approved drug may add to recruitment problems rather than diminish them by removing one of the incentives that private practitioners now have to refer patients to a clinical trial. We may see more early presentations for care but fewer research subjects.
Acute stroke patients are already being treated with available agents. The "unapproved use of an approved agent" outside of a trial is a disturbing issue. Under Food and Drug Administration regulations, a physician may treat patients with an approved drug for indications not included on the approved drug label. Physicians are increasingly tempted to use thrombolysis for stroke "off protocol" or under an in-house protocol approved by an institutional review board. This obviates the need to participate in a controlled trial and circumvents trial methodology (including randomization and placebo treatment, with which many physicians are uncomfortable) and slows the completion of carefully designed multicenter trials by reducing patient accrual.
Participation in acute stroke treatment trials is arduous, carries little academic recognition unless the investigator is involved in publication of trial results, and may not provide sufficient financial compensation to protect the time of clinical investigators or ensure the support of stroke fellows and a dedicated research staff. Per capita payment, now the norm, means that smaller centers with erratic enrollment cannot maintain a stroke research team without other support. The costs for advertising on "prime-time" television and other media to promote stroke awareness, use of emergency services, and drug treatment availability (now beyond what individual stroke programs can afford and study sponsors are willing to underwrite) could be shared among several support sources. Fewer better-financed trials might encourage regional cooperation rather than competition, enhance research access to emergency medical systems, provide resources for high-impact patient education, and support well-organized and motivated research teams. A more cooperative approach might also have advantages for pharmaceutical firms, which could concentrate on developing convincing preclinical and phase 1 and 2 data in anticipation of being part of a phase 3 study in which the financial exposure of an individual company might be reduced.
Stroke trial design issues are centrally involved with the agents to be tested. As with fibrinolysis in myocardial infarction and nitrogen mustard for lymphoma, proof of efficacy of a single agent before testing combination therapy has been taken as the standard design approach. It has been unclear whether the lack of efficacy of any treatment for acute stroke represents intrinsic lack of efficacy, study design flaws, or both. A false-negative study may doom a potentially therapeutic drug. In the setting of acute stroke, the financial resolve of a sponsor is also an important variable.
The complexity of focal cerebral ischemia, its clinical variability, and the lack of widely applicable methods to rapidly distinguish pathophysiological subtypes raise additional issues regarding study design. It has been suggested that the complex pathophysiology of cerebral ischemia6 7 8 may require a multiple, simultaneous, or sequential drug-treatment approach.9 A variety of processes occur at the cellular level that may ultimately lead to irreversible injury and cell death, including early loss of ion homeostasis, presynaptic release of excitatory neurotransmitter, intracellular calcium accumulation and acidosis, formation of free radicals, nitric oxide synthesis, adhesion and recruitment of polymorphonuclear leukocytes, and other as yet unidentified processes. Therapy directed at only one of the processes involved in focal brain ischemia may have only modest benefits at best,9 10 and even these may be obscured by inadequate sample sizes or overwhelmed by the complexity of acute ischemia. The concern here is that a series of single-agent trials might reject agents of limited efficacy in isolation that are effective in combination with other drugs. Testing of single agents is a logical first step. However, the first modern randomized clinical trial was in fact a test of single and combined therapy in pulmonary tuberculosis, and it showed that two drugs were more effective than bed rest or either drug alone.11 12 In cerebral ischemia, combination therapy may produce enhanced salvage of ischemic tissue and improve clinical outcome. The paradigm of "sponsor-driven," single-drug trials, however, does not readily lend itself to testing this approach unless one company controls the proprietary rights to multiple drugs.
Factorial study designs typically combine drugs that are already approved for the indication of interest or are on the market for other uses. Such a design has been used in some stroke-prevention studies but as yet has been attempted in only two acute stroke trials.13 Advantages include the testing of combinations of agents. Disadvantages include the complexity of design, larger sample-size requirements, and the need for data on combined toxicity. New drugs could be tested in combination, provided sufficient data on safety were available. With this approach, given adequate sample size, the efficacy of two agents may be evaluated against either agent alone or no treatment. In some versions of this design, relative toxicity data could be obtained as part of the trial.
Finally, for any trial approach the suitability of clinical outcome measures, preferably validated with prospective studies, is highly relevant. The extent to which potential surrogate measures such as imaging techniques14 could be used to assess efficacy is a crucial related issue. How outcomes are measured has a real impact on study size and success.
What is needed is a national strategic plan for stroke that incorporates clinical research needs within a healthcare environment that optimizes acute stroke care delivery. We propose that the National Institutes of Neurological Diseases and Stroke convene a consensus forum or workshop to address critical issues of trial design, sponsorship, recruitment, and execution, as well as a consideration of the healthcare delivery environment in which these trials must take place. Participants could include academic scientists and institutions; managed-care, private-practice, industry, regulatory, and scientific federal agencies; large medical payers; the American Heart and National Stroke Associations; professional medical societies (including primary-care and emergency medicine and technician groups); and patient-advocacy groups. The growing influence of large payers should be examined, since these forces may either create incentives that could encourage trial participation or adopt policies that make this research much more difficult. Many of the issues we raise here will continue to be relevant even if currently ongoing trials produce an approved initial treatment for some stroke patients.
A broader goal could be an attempt to guide the "market forces" that are driving changes in medical-care delivery in the United States today to ensure the viability of clinical stroke treatment trials and facilitate access to expert care for stroke patients. We believe that such an effort could be important. During the evolution of organized and advanced trauma care in the United States, the recommendations of bodies such as the American College of Surgeons/Committee on Trauma and the American College of Emergency Physicians have helped shape the way these important services are now provided by proposing standards of care and competency for trauma centers.15 16 A similar systematic approach, perhaps organized on a regional basis and designed with an appreciation for the importance of clinical research in mind, should be considered for the care of the acute stroke patient. Such an undertaking would be difficult and certainly not without controversy. However, we believe that it would be worth the effort and could result in more effective strategies for testing new treatments for acute ischemic stroke and the provision of optimal acute care for our patients with stroke.
|
References |
|---|
|
TopIntroductionReferences |
|---|
1. Adams HP. Trials of trials in acute ischemic stroke. Stroke. 1993;24:1410-1415.
2.
Barsan WG, Brott TG, Broderick JP, Haley EC, Levy DE,
Marler JR. Time of hospitalization in patients with acute
stroke. Arch Intern Med. 1993;153:2558-2561.
3. Barsan WG, Brott TG, Olinger CP, Marler JR. Early treatment for acute ischemic stroke. Ann Intern Med. 1989;111:449-451.
4.
Saver JL. Coping with an embarrassment of
riches: how stroke centers may participate in multiple, concurrent
clinical stroke trials. Stroke. 1995;26:1289-1292.
5.
Marwick C. New panel tackles clinical research funding
cuts. JAMA. 1995;274:526.
6. Pulsinelli W. Pathophysiology of acute ischemic stroke. Lancet. 1992;339:533-536. [Medline] [Order article via Infotrieve]
7. Siesjö BK. Pathophysiology and treatment of focal brain ischemia. J Neurosurg. 1992;77:169-184, 337-354. [Medline] [Order article via Infotrieve]
8. Jones TH, Morawetz RB, Crowell RM, Marcoux FW, FitzGibbon SJ, DeGirolami U, Ojemann RG. Thresholds of focal cerebral ischemia in awake monkeys. J Neurosurg. 1981;54:773-782. [Medline] [Order article via Infotrieve]
9.
Hallenbeck JM, Frerichs KU. Stroke therapy: it
may be time for an integrated approach. Arch Neurol. 1993;50:768-770.
10. Grotta J. The current status of neuronal protective therapy: why have all neuronal protective drugs worked in animals but none so far in stroke patients? Cerebrovasc Dis. 1994;4:115-120.
11. Daniel M, Hill AB. Chemotherapy of pulmonary tuberculosis in young adults. Brit Med J. 1952;1:1162-1168.
12. Ingelfinger FJ. The randomized clinical trial. N Engl J Med. 1972;287:100-101.
13. Major Ongoing Stroke Trials. Stroke. 1995;26:349-353.
14.
Hasegawa Y, Fisher M, Latour LL, Dardzinski BJ, Sotak
CH. MRI diffusion mapping of reversible and irreversible
ischemic injury in focal brain ischemia.
Neurology. 1994;44:1484-1490.
15. American College of Surgeons Committee on Trauma. Resources for Optimal Care of the Injured Patient. Chicago, Ill: ACS; 1993.
16. American College of Emergency Physicians. Guidelines for trauma care systems. Ann Emerg Med. 1987;16:459-463.[Medline] [Order article via Infotrieve]
This article has been cited by other articles:
 |
|
 |
|
  S. R. Levine and M. Gorman "Telestroke" : The Application of Telemedicine for Stroke Stroke, February 1, 1999; 30(2): 464 - 469. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
M. S. Pessin, H. P. Adams Jr, R. J. Adams, M. Fisher, A. J. Furlan, W. Hacke, E. C. Haley, M. F. Hazinski Jr, C. M. Helgason, R. T. Higashida, et al. Acute Interventions Stroke, July 1, 1997; 28(7): 1518 - 1521. [Full Text]
|
|
|
|
|
|
P. J. Dorman and P. A.G. Sandercock Considerations in the Design of Clinical Trials of Neuroprotective Therapy in Acute Stroke Stroke, September 1, 1996; 27(9): 1507 - 1515. [Full Text]
|
|
| |||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||