With a Little Help From My Friends
2017; Lippincott Williams & Wilkins; Volume: 48; Issue: 8 Linguagem: Inglês
10.1161/strokeaha.117.017244
ISSN1524-4628
Autores Tópico(s)Cerebral Venous Sinus Thrombosis
ResumoHomeStrokeVol. 48, No. 8With a Little Help From My Friends Free AccessReview ArticlePDF/EPUBAboutView PDFView EPUBSections ToolsAdd to favoritesDownload citationsTrack citationsPermissions ShareShare onFacebookTwitterLinked InMendeleyReddit Jump toFree AccessReview ArticlePDF/EPUBWith a Little Help From My FriendsSeeking Consistent and Persuasive Evidence E. Clarke HaleyJr, MD E. Clarke HaleyJrE. Clarke HaleyJr From the Department of Neurology, University of Virginia Health System, Charlottesville. Originally published16 Jun 2017https://doi.org/10.1161/STROKEAHA.117.017244Stroke. 2017;48:2326–2328Other version(s) of this articleYou are viewing the most recent version of this article. Previous versions: January 1, 2017: Previous Version 1 "I get by with a little help from my friends."— The Beatles, 19671The past 30 years have seen remarkable advances in all areas of medicine, but perhaps none more so than stroke. On the 10th anniversary of the passing of Dr David Sherman, scholar, compassionate clinician, leader, family man, and friend, it is appropriate to consider the progress we have made in large part from his leadership. The care of patients with stroke was much different in 1984 when I joined the faculty of the School of Medicine at the University of Virginia. Stroke was still the third leading cause of death right behind heart disease and cancer. Patients with stroke were generally cared for by primary care physicians, and neurologists and neurosurgeons were largely adnexal. There was no approved acute treatment. For neurologists, our main job was to consult the next day to localize the lesion, provide some help with prognosis, and help decide who should be on heparin infusions. Although computed tomography was widely available by then, MRI was still in development. We had carotid ultrasound, and transcranial Doppler was just being introduced, but the only reliable way to see the cerebral vasculature was with catheter angiography with its attendant risks. Warfarin was widely used for stroke prevention but not for atrial fibrillation. The only proven stroke preventative, aside from risk factor control, was aspirin.Fast forward to now, and what a different world in which we live! Stroke has slipped to fifth from third on the list of causes of death of Americans.2 Stroke teams, stroke units, and stroke centers led by specialists are widely distributed. Intravenous tissue-type plasminogen activator (tPA) is approved for treatment of patients with acute stroke within 3 hours of onset and is routinely administered to patients with acute stroke within 4.5 hours from onset. There are now endovascular treatments for extra- and intracranial occlusive disease, intracranial aneurysms, and arteriovenous malformations. A new generation of oral anticoagulants is approved for stroke prevention in atrial fibrillation. Noninvasive imaging of the entire intra- and extracranial vasculature is now routine. Confirmation of an ischemic stroke diagnosis is available immediately with diffusion-weighted MRI.And like the song says, "the future's so bright, I gotta wear shades!"3 Just a few months ago, Congress passed, and the President signed into law, the 21st Century Cures Act.4 This optimistic legislation promises to accelerate the discovery, development, and delivery of 21st Century Cures by, among other things, injecting >$5 billion into National Institutes of Health–sponsored research, as well as stripping away what some perceive as excessive red tape associated with the approval process for new drugs and devices. Like always, though, the devil is in the details, and certain provisions of the Act, particularly as they apply to the United States Food and Drug Administration (FDA), are controversial. Among other directives, the FDA is mandated to establish processes under which patient experience data may be considered in the risk-benefit assessment of a new drug, issue guidance that addresses using alternative statistical methods in the development and review of drugs, and to evaluate the use of evidence from clinical experience (in place of evidence from clinical trials) in supporting approval of a drug for a new indication. Although it remains to be seen how these mandates will actually be enacted at the regulatory level, some have raised concern that the bar for approval of new drugs and devices (or old drugs for new indications) has been substantially lowered, thereby raising the risk that ineffective treatments could be approved.5 The debate over these provisions has been heated and contentious (what is not anymore?). However, might the stroke experience shed some light on this debate?In August 1995, the Steering Committee of the National Institute for Neurological Diseases and Stroke (NINDS) tPA Stroke Study Group gathered at the Airlie Mansion in rural Fauquier County Virginia. Dr John Marler, our NINDS Project Officer, and Dr Barbara Tilley, the director of the trial's Statistical and Data Management Center, were in attendance along with Drs Don Easton and Mike Walker, representing our Data and Safety Monitoring Board (DSMB). At this meeting, Barbara presented the raw results of the NINDS tPA Stroke Trial, both parts 1 and 2, to the assembled investigators and co-ordinators who had worked so hard for the previous 5 years (some of us almost 10!) to bring the trial(s) to a successful conclusion. Needless to say, after her presentation, we could not have been any happier! What most people do not know, though, is how close we came to not having this moment.Flash back to the summer of 1992. Despite the relatively small number of centers and the severe time restrictions under which we were operating (ie, randomizing and treating stroke patients within 90 and 180 minutes from onset), recruitment was running ahead of schedule. By that summer, we were actually getting close to completing enrollment in what became to be known as part 1 of the trial. Part 1 was originally designed as a stand-alone, middle phase, proof of concept (so-called phase 2B) trial to demonstrate the clinical activity of early administration of tPA in patients with acute stroke, as measured by the proportion of patients who showed dramatic neurological improvement within 24 hours. Because there was no consensus at the time on what a long-term efficacy end point for stroke should be, we collected 4 of them to help us design the next trial if this first one was successful. The fly in the ointment, though, was that those of us in the trenches were regularly receiving phone calls from our colleagues in emergency medicine or neurology at other hospitals, all relating variations on the following scenario:"Dr. Haley, I have this 88 year old woman in the Emergency Room who had the witnessed onset of aphasia and a right hemiparesis about an hour ago. She's the mother of the mayor and still is very active and lives independently. Her head CT is negative. She has a lot of family members here, including the mayor, and they are naturally quite upset at this sudden turn of events, and want 'everything' to be done. I know that you guys are doing a clinical trial of tPA, but you are too far away to send her to you. What's the dose you are using?"Of course, tPA had already been approved for myocardial infarction and was readily available off the shelf at virtually every hospital. We naively surmised that with no proven treatment for stroke available, if we were to stick to our original plan to complete the phase 2B, analyze and report the results, and then design and execute the definitive efficacy trial, if the phase 2B study was positive, we would never get a chance to complete the efficacy trial because everyone would demand tPA. Perhaps further exacerbating the problem, a couple of small, underpowered randomized trials had previously suggested benefit,6,7 and a recently published meta-analysis had also suggested benefit.8So amid all of this collective indigestion, we decided, after considering all our options, to propose a protocol amendment to the FDA and our DSMB that would allow us to complete enrollment in part 1, but to design the efficacy trial now without any unblinding of the data to the investigators, share the results of part 1 with our DSMB, and if they were encouraging, roll seamlessly into a second trial, part 2, with 3-month outcome as the primary end point. If part 1 was not encouraging, we would stop.The DSMB met with us on December 4, 1992. After prolonged open and closed sessions, they gathered us together and advised that they had approved our protocol modification and that we still had an ethical trial and should continue to complete enrollment and follow-up in part 1. In the interim, we should design part 2 in such a way as to provide "consistent and persuasive evidence" that tPA was efficacious and that the benefits outweighed the risks. In the interval, there was to be no unblinding of any of the investigators.Thus, we had our marching orders. With regard to the primary end point, we had fairly good knowledge that a 3-month outcome would be acceptable to the FDA although we would continue to follow patients out to a year. The outcome had to focus not just on improvements in the neurological examination but on actual functional outcome and differences that were clinically meaningful not just to neurologists but to patients. We chose to focus on the proportion of patients who had either complete or near complete recovery on their neurological examination, activities of daily living, and residual handicap. What kept us awake at night was worrying about what would happen if we chose the wrong measure as the primary end point, and it failed to meet statistical significance but was trending in the right direction while all the other end points were statistically significantly better in the tPA-treated group. To address this issue, NINDS held a workshop in Bethesda in September 1993 and thus was adopted the global outcome statistic, which combined the results of all 4 of our outcome scales to give 1 unified statistic that took into account the intercorrelated nature of the scales, but would not make a silk purse out of a sow's ear if there was not consistent improvement among all the scales.9Back to 1995: as it turns out, we need not have worried. The results of part 2, the efficacy trial, are well known.10 The difference in each of the 4 scales (the National Institutes of Health Stroke Scale, the Barthel Index, the modified Rankin Scale, and the Glasgow Outcome Scale) was statistically significant (P<0.05) in its own right, with absolute treatment differences in favor of tPA ranging from 11 to 13 percentage points. Combining them into the global outcome statistic only made the results more significant (P 4, we would have achieved statistical significance.11 Thank goodness our DSMB had the good judgment to recognize that and not stop our trial in 1992!So the results of both part 1 and part 2 were resoundingly positive. Even with relatively small sample sizes, the results in favor of tPA treatment were both statistically significant and clinically meaningful. In June 1996, the FDA approved tPA for stroke within 3 hours of onset. Guidelines from the American Stroke Association, citing level 1 evidence, and the American Academy of Neurology recommending its use for properly selected patients, followed shortly thereafter.12,13 At last, in the words of my friend, Dr Harold Adams, in accepting his award from this organization in 1992, "We have a treatment for stroke, and we know it!"14 Or did we?Sadly, our optimism was not quickly justified. Despite aggressive education campaigns by the manufacturer and NINDS, absolutely no increase in use was appreciated during the first 2 years after approval. By the spring of 1998, only 1% of all patients with stroke were being treated, and only 2% of all time-eligible patients were getting treatment.15 If tPA for stroke was a tree falling in the forest with no one listening, it did make a sound and that sound was thud!What happened? The reasons for the slow adoption of tPA for stroke are multifactorial and have been recounted elsewhere.16 Although some inertia is expected with any new treatment that requires, in some cases, massive reorganization around how emergency services are provided at individual hospitals, there is no doubt that a concerted effort was mounted to discredit the results of the NINDS trials, and in some cases its investigators, government sponsors, and even the volunteer membership of guideline writing committees.17 The data from the trials were reanalyzed over and over again by different interest groups with a new set of conclusions, some might argue pre-determined, produced each time. I would posit that there has perhaps been no data set in medical history that has undergone more scrutiny than the NINDS tPA Stroke Trial.Yet the data held up!18 The consistent and persuasive evidence demanded by our DSMB in 1992 finally prevailed. After about a decade, use of tPA for stroke slowly began to increase. Schwamm et al19 reported that based on Get With the Guidelines data, by 2011, tPA use had increased to 7% of all stroke and >70% of patients who were time eligible, the implication being that 1000s more Americans were now enjoying complete or near complete recovery from their strokes and that the healthcare system is saving even more money from the long-term costs of stroke care. Hopefully, those numbers are continuing to increase.So, let this be a cautionary tale to clinicians, investigators, manufacturers, and regulators alike. Although there is certainly a role for patient experience data, novel statistics, and observational research in a variety of circumstances, when it is really crunch time, we must insist on consistent and persuasive evidence that in most cases can only be provided by properly designed and executed randomized, concurrently controlled, and as much as possible, blinded clinical trials. Otherwise, we and our patients may be left twisting in the wind when trying to consider using a new treatment.AcknowledgmentsI thank the International Stroke Conference Program Committee for making me its Sherman Awardee for 2017. I am also grateful to many others, too numerous to name, who provided mentoring, advice, inspiration, and support to me throughout my career in stroke.DisclosuresNone.FootnotesThe David G. Sherman Lecture was presented at the International Stroke Conference, Houston, TX, February 22, 2017.Correspondence to E. Clarke Haley, Jr, MD, Department of Neurology, University of Virginia Health System, Box 800394, Charlottesville, VA 22908. E-mail [email protected]References1. Lennon J, Paul M. Sgt. Pepper's Lonely Hearts Club Band. EMI Records, 1967. CD.Google Scholar2. Kochanek KD, Murphy SL, Xu J, Tejada-Vera B. Deaths: final Data for 2014.Natl Vital Stat Rep. 2016; 65:1–122.MedlineGoogle Scholar3. Pat M. Greetings from Timbuk3. Timbuk3. I.R.S., 1986. CD.Google Scholar4. 21st Century Cures Act, Pub L No.114–255, 133 Stat 1033.Google Scholar5. Avorn J, Kesselheim AS. The 21st century cures act–will it take us back in time?N Engl J Med. 2015; 372:2473–2475. doi: 10.1056/NEJMp1506964.CrossrefMedlineGoogle Scholar6. Mori E, Yoneda Y, Tabuchi M, Yoshida T, Ohkawa S, Ohsumi Y, et al. Intravenous recombinant tissue plasminogen activator in acute carotid artery territory stroke.Neurology. 1992; 42:976–982.CrossrefMedlineGoogle Scholar7. Haley EC, Brott TG, Sheppard GL, Barsan W, Broderick J, Marler JR, et al. Pilot randomized trial of tissue plasminogen activator in acute ischemic stroke. The TPA Bridging Study Group.Stroke. 1993; 24:1000–1004.LinkGoogle Scholar8. Wardlaw JM, Warlow CP. Thrombolysis in acute ischemic stroke: does it work?Stroke. 1992; 23:1826–1839.LinkGoogle Scholar9. Tilley BC, Marler J, Geller NL, Lu M, Legler J, Brott T, et al. Use of a global test for multiple outcomes in stroke trials with application to the National Institute of Neurological Disorders and Stroke t-PA Stroke Trial.Stroke. 1996; 27:2136–2142.LinkGoogle Scholar10. The National Institute of Neurological Disorders and Stroke rt-PA Stroke Study Group. Tissue plasminogen activator for acute ischemic stroke.New Engl J Med1995; 333:1581–1587.CrossrefMedlineGoogle Scholar11. Haley EC, Lewandowski C, Tilley BC. Myths regarding the NINDS rt-PA Stroke Trial: setting the record straight.Ann Emerg Med. 1997; 30:676–682.CrossrefMedlineGoogle Scholar12. Adams HP, Brott TG, Furlan AJ, Gomez CR, Grotta J, Helgason CM, et al. Guidelines for thrombolytic therapy for acute stroke: a supplement to the guidelines for the management of patients with acute ischemic stroke. A statement for healthcare professionals from a special writing group of the Stroke Council, American Heart Association.Stroke. 1996; 27:1711–1718.MedlineGoogle Scholar13. Greenberg MK, Alter M, Ashwal S, Calverly J, Franklin G, French J, et al; for the American Academy of Neurology Quality Standards Subcommittee. Practice advisory. Thrombolytic therapy for acute ischemic stroke – summary statement.Neurology1996; 47:835–839.CrossrefMedlineGoogle Scholar14. Adams HP. Trials of trials in acute ischemic stroke. The Humana Lecture.Stroke. 1993; 24:1410–1415.LinkGoogle Scholar15. Hacke W, Brott T, Caplan L, Meier D, Fieschi C, von Kummer R, et al. Thrombolysis in acute ischemic stroke: controlled trials and clinical experience.Neurology. 1999; 53(7 suppl 4):S3–S14.MedlineGoogle Scholar16. Lyden P. Why don't more patients receive intravenous rt-PA for acute stroke?Expert Rev Neurother. 2015; 15:571–574. doi: 10.1586/14737175.2015.1041510.CrossrefMedlineGoogle Scholar17. Lenzer J. Alteplase for stroke: money and optimistic claims buttress the "brain attack" campaign.BMJ. 2002; 324:723–729.CrossrefMedlineGoogle Scholar18. Ingall TJ, O'Fallon WM, Asplund K, Goldfrank LR, Hertzberg VS, Louis TA, et al. Findings from the reanalysis of the NINDS tissue plasminogen activator for acute ischemic stroke treatment trial.Stroke. 2004; 35:2418–2424. doi: 10.1161/01.STR.0000140891.70547.56.LinkGoogle Scholar19. Schwamm LH, Ali SF, Reeves MJ, Smith EE, Saver JL, Messe S, et al. Temporal trends in patient characteristics and treatment with intravenous thrombolysis among acute ischemic stroke patients at Get With The Guidelines-Stroke hospitals.Circ Cardiovasc Qual Outcomes. 2013; 6:543–549. doi: 10.1161/CIRCOUTCOMES.111.000303.LinkGoogle Scholar Previous Back to top Next FiguresReferencesRelatedDetailsCited ByGoldstein L (2018) Twenty-First Century Cures Act, Stroke, 49:10, (2555-2558), Online publication date: 1-Oct-2018. August 2017Vol 48, Issue 8 Advertisement Article InformationMetrics © 2017 American Heart Association, Inc.https://doi.org/10.1161/STROKEAHA.117.017244PMID: 28626047 Manuscript receivedApril 5, 2017Manuscript acceptedMay 22, 2017Originally publishedJune 16, 2017Manuscript revisedMay 10, 2017 KeywordsUnited States Food and Drug Administrationstroketissue plasminogen activatorPDF download Advertisement SubjectsIschemic StrokeTreatment
Referência(s)