On Sept. 25, 2020, ACEP Now published Ken Milne’s article entitled “After Re-Analysis, No Trials Show Efficacy of tPA in Acute Ischemic Stroke,” which focuses on reanalysis of NINDS (0–3 hours post-stroke onset) and ECASS III (3–4.5 hours post-stroke onset) data.

Excerpt

A [graphical] reanalysis of the NINDS data published in 2009 revealed that a baseline imbalance in stroke severity at presentation likely led to the difference in outcomes.¹ After controlling for these baseline differences, the claimed efficacy of tPA was no longer statistically significant.

Concern 1: Two responses to the reanalysis were published. Saver et al indicate this reanalysis of NINDS “depart[s] from best practices appropriate for the visual display of quantitative information.”² They continue on to say, “Several methods exist that are appropriate to the graphical depiction of scales with ordinal functional values and skewed population distributions, including charting normalized gain and loss and charting clinically relevant ordinal categories. Graphical analysis of the NIHSS and delta NIHSS scores in the two NINDS-TPA trials, when conducted in this proper manner, delineate large magnitude treatment benefits of under 3 hour fibrinolytic therapy in acute stroke.”

In a second response to this reanalysis, Dewey et al point out “such a conclusion cannot be justified by the findings from this post-hoc analysis of a secondary outcome from a randomized controlled trial with a positive primary outcome.”³ They also “question the relevance of the NIHSS as a measure of outcome at 90 days. At this late time point functional capacity is of much more direct relevance to patients and is routinely assessed with the modified Rankin and Barthel Scales.”

As noted by these two different responses, the reanalysis was performed on a secondary outcome and did not account for the ordinal, noninterval nature of NIHSS’ functional significance and the skewed population distribution, leading to misrepresentation of the results from the NINDS trial. The author also failed to discuss the graphical reanalysis of the two NINDS-tPA trials completed by Saver et al.⁴

Excerpt

Reanalysis of the ECASS III trial data with multiple approaches adjusting for baseline imbalances does not support any significant benefits and continues to support harms for the use of alteplase 3–4.5 hours after stroke onset.

Concern 2: Although Alper et al concluded their reanalysis of ECASS III data did not demonstrate significant benefits, they do note that a “limitation of reanalysis, or any method for adjusting for non-randomised factors influencing the effect estimates from a randomised trial, is such analyses cannot confidently produce new conclusions (neither a claim of efficacy nor a claim of absence of efficacy). … In this case the reanalysis does not negate the original findings, but it greatly reduces the certainty for those findings.”⁵

Although not obligatory, the author does not call attention to the flaws of conducting this reanalysis. As Activase in the 3–4.5-hour time frame is not FDA-approved, Genentech does not endorse the use of Activase use in this time frame. However, a reanalysis that examines the benefit or risk of a society-recommended treatment should be presented objectively to allow informed decision making on whether or not to offer such treatment to patients.

Emergency physicians rely on professional organizations such as ACEP to provide data driven information to help inform clinical practice.

Excerpt

There are 13 RCTs of thrombolytics for AIS (see Table 1). Four were stopped early for harm (bleeding) or futility, and all 13 failed to show a statistical benefit after the reanalysis of NINDS-2 and ECASS-III.

Concern 3: Table 1 summarizes RCTs evaluating various thrombolytics for treatment of acute ischemic stroke. Because this table includes results from streptokinase and desmoteplase trials and from trials allowing time of alteplase treatment up to 6 hours from onset, inappropriate conclusions may be made about the safety and efficacy of alteplase for treatment of acute ischemic stroke.

Activase was FDA-approved for the treatment of acute ischemic stroke within 3 hours of symptom onset, based on the data from the two-part NINDS trial.¹⁶ While symptomatic intracranial hemorrhage (sICH) was significantly higher in the Activase group (6.4 percent) versus placebo (0.6 percent), mortality was lower in the Activase arm (17.3 percent) compared to the placebo arm (20.5 percent).

FDA Clearance:

• Activase is not FDA-approved for use in treatment of acute ischemic stroke 3–4.5 hours from onset.
• Activase is FDA-approved for the treatment of acute ischemic stroke within 3 hours from onset of symptoms.
• Please refer to the product prescribing information for the full FDA-approved indications and safety information, available at www.gene.com/download/pdf/activase_prescribing.pdf.

Emergency physicians rely on professional organizations such as ACEP to provide data-driven information to help inform clinical practice. Physicians are then responsible for evaluating the data and using their best judgment for incorporation into practice. This analysis requires presentation of unbiased material with each side impartially represented.

Respectfully yours,

Rachel Garvin, MD, Senior Medical Director, Lytics, Genentech, Emergency Physician and Neuro Critical Care Specialist

Michael Liberman, MD, Senior Group Medical Director, Lytics US Medical Affairs, Genentech

References

1. Hoffman JR, Schriger DL. A graphic reanalysis of the NINDS trial. Ann Emerg Med. 2009;54(3):329-336.
2. Saver JL, Gornbein J, Starkman S. Response to: “A graphic reanalysis of the NINDS Trial.” Ann Emerg Med. 2010;55(2):226-227; author reply 229.
3. Dewey HM, Churilov L, Blacker D, et al. Response to “A graphic reanalysis of the NINDS Trial.” Ann Emerg Med. 2010;55(2):227-229; author reply 229.
4. Saver JL, Gornbein J, Starkman S. Graphic reanalysis of the two NINDS-tPA trials confirms substantial treatment benefit. Stroke. 2010;41(10):2381-2390.
5. Alper BS, Foster G, Thabane L, et al. Thrombolysis with alteplase 3-4.5 hours after acute ischaemic stroke: trial reanalysis adjusted for baseline imbalances. BMJ Evid Based Med. 2020;25(5):168-171.
6. Activase [package insert]. South San Francisco, CA: Genentech, Inc.; 2018.

The Authors Respond

We are grateful for the letter by Garvin and Liberman, employees of Genentech. Our ACEP Now article was focused on the 2020 reanalysis of ECASS III by Alper et al, which added to the uncertainty about stroke thrombolysis. The 1995 NINDS trials were background information on the uncertainty.^1,2

For a treatment to be scientifically sound, there must be replication of studies, minimization of bias, and healthy debate. These requirements have not been met for stroke thrombolysis. Discussions on controversies contribute to our understanding of deficiencies in existing data. The most glaring issue on this subject is the absence of replication studies suggesting benefit. The medical literature is replete with initially “positive” studies followed by larger, more reliable “negative” studies. This has contributed to the phenomenon of medical reversal whereby newer, superior results contradict current practice.³ There are many examples where society guidelines promoting harmful treatments (eg, steroids for spinal cord injury) were later overturned.⁴

The concerns about the National Institutes of Health Stroke Scale (NIHSS) being a nonlinear scale are recognized. These concerns, although theoretically valid, do not appear to be relevant to NINDS, as the delta NIHSS was the same for all the treatment arms at every area of the scale—it changed equally with (early or later) tPA as with placebo for small strokes, for moderate ones, and for severe strokes. Changes in NIHSS measure discrete elements of neurological function rather than the more important overall functional status. Still, we might ask, “just how did tPA lead to better overall outcomes … if it had no effect on any element of neurological function?”⁵

Alper et al also raised concerns around the dangers of selective analysis of trials. They refer to this as “selective analysis reporting bias” and state this could be involved in the stroke literature, specifically the reporting of ECASS III.⁶ In reality, reanalyses are not often done. There is evidence that, when they are conducted, they frequently change the direction and magnitude of effect size and statistical significance.⁷

No claim was made in the ACEP Now article as to which analysis was correct. The bottom line stated that the reanalysis by Alper et al does not support a patient-oriented benefit of tPA given 3–4.5 hours after onset of stroke symptoms and confirms the known potential harm. This agrees with the conclusions of Alper et al’s reanalysis, and we echo their call for the readers to reconsider the use of tPA in this time window.

Excerpt

Conclusions: Reanalysis of the ECASS III trial data with multiple approaches adjusting for baseline imbalances does not support any significant benefits and continues to support harms for the use of alteplase 3–4.5 hours after stroke onset. Clinicians, patients, and policymakers should reconsider interpretations and decisions regarding management of acute ischaemic stroke that were based on ECASS III results.

tPA is not FDA-approved in the 3–4.5-hour time frame. However, the 2015 ACEP policy does discuss the use of tPA in this time window. The recommendations were stated in the ACEP Now article background section and included a link to the full ACEP policy statement.⁸ This highlighted the importance of shared decision making about the potential benefits and potential harms of tPA in this time frame.

The table in our article provides some details on 13 foundational randomized control trials (RCTs) evaluating thrombolysis for acute ischemic stroke. Each thrombolytic agent is clearly identified. There is no definitive data to suggest one thrombolytic agent is better than another.⁹ To selectively highlight two “positive” tPA trials (both with questionable efficacy on reanalysis) and omit the other 11 RCTs (six tPA and five non-tPA) that did not report benefit for their primary outcome is a form of the  “Texas sharpshooter fallacy” (ie, “counting” the positive trials but ignoring the negative ones).¹0 Or putting it differently, all the evidence is positive, as long as we reject any evidence that fails to be positive.

For a treatment to be scientifically sound, there must be replication of studies, minimization of bias, and healthy debate. These requirements have not been met for stroke thrombolysis.

It is important to note that there have been no “positive” replication studies. In addition, two tPA trials were stopped early due to harm or futility.^11,12 Importantly, it is recognized that harms are underreported in RCTs, systematic reviews, and meta-analysis (SRMA).^13,14

Garvin and Liberman are correct that symptomatic intracranial hemorrhage (sICH) was higher in the tPA group and mortality was lower compared to placebo. However, they failed to mention the statistical result. The harm (5.8 percent absolute increase in sICH) was statistically significant (P <0.001) while the benefit (absolute decrease in mortality 3.2 percent) was not (P=0.3). Shinton raised concerns regarding the NINDS methodology, stating “the evidence of benefit is precarious.”¹⁵ Overall, there is more confidence in the increase in harm (bleeding) than the potential decrease in mortality. In fact, two recent SRMA have reported a statistical increase in early mortality with tPA and a nonstatistical increase in late mortality.^16,17

We note that Garvin and Liberman introduce the idea of being unbiased and impartial. The potential financial conflicts of interest (COI) around tPA have been documented and are a powerful form of bias.¹⁸ It is also known that COI can introduce bias into RCTs and SRMAs and need to be managed during guideline drafting.^19–21 It is ironic that two Genentech employees suggest we are biased when there has been so much pro-tPA promotion in major journals. Indeed, in “Everyone’s a Little Biased (Even Physicians),” Cain and Detsky point out that “everyone is likely capable of rationalizing beliefs and denying influences that bias them. The most important action physicians can take as a profession is to recognize this.”²²

We agree that physicians are responsible for evaluating the data and using their best judgment for incorporation into practice. We encourage physicians to read and critically appraise the primary literature, reflect upon their clinical experience, and engage with patients about their values and preferences. This is the foundation of evidence-based medicine.

Respectfully,

Ken Milne, MSc, MD, CCFP(EM), FCFP, FRRMS, Schulich School of Medicine and Dentistry, Western University

Daniel M. Fatovich, MBBS, FACEM, PhD, professor of emergency medicine, University of Western Australia; head of the Centre for Clinical Research in Emergency Medicine, Harry Perkins Institute of Medical Research 

References

1. Alper BS, Foster G, Thabane L, et al. Thrombolysis with alteplase 3-4.5 hours after acute ischaemic stroke: trial reanalysis adjusted for baseline imbalances. BMJ Evid Based Med. 2020;25(5):168-171.
2. National Institute of Neurological Disorders and Stroke rt-PA Stroke Study Group. Tissue plasminogen activator for acute ischemic stroke. N Engl J Med. 1995;333(24):1581-1587
3. Prasad V, Cifu A. Medical reversal: why we must raise the bar before adopting new technologies. Yale J Biol Med. 2011;84(4):471-478.
4. Lenzer J. Why we can’t trust clinical guidelines. BMJ. 2013;346:f3830.
5. Fatovich DM. Believing is seeing: stroke thrombolysis remains unproven after the third International Stroke Trial (IST-3). Emerg Med Australas. 2012;24(5):477-479.
6. Alper BS. The dangers of selective analysis: has stroke treatment been misguided for a decade? BMJ website. Accessed March 26, 2021.
7. Ebrahim S, Sohani ZN, Montoya L, et al. Reanalyses of randomized clinical trial data. JAMA. 2014;312(10):1024-1032.
8. Brown MD, Burton JH, Nazarian DJ, et al. Clinical policy: use of intravenous tissue plasminogen activator for the management of acute ischemic stroke in the emergency department. Ann Emerg Med. 2015;66(3):322-333.
9. Wardlaw JM, Koumellis P, Liu M. Thrombolysis (different doses, routes of administration and agents) for acute ischaemic stroke. Cochrane Database Syst Rev. 2013(5):CD000514.
10. The Texas Sharpshooter. Your Logical Fallacy Is website. Accessed March 26, 2021.
11. Clark WM, Wissman S, Albers GW, et al. Recombinant tissue-type plasminogen activator (alteplase) for ischemic stroke 3 to 5 hours after symptom onset. The ATLANTIS Study: a randomized controlled trial. Alteplase Thrombolysis for Acute Noninterventional Therapy in Ischemic Stroke. JAMA. 1999;282(21):2019-2026.
12. Clark WM, Albers GW, Madden KP, et al. The rtPA (alteplase) 0- to 6-hour acute stroke trial, part A (A0276g): results of a double-blind, placebo-controlled, multicenter study. Thromblytic therapy in acute ischemic stroke study investigators. Stroke. 2000;31(4):811-816.
13. Hodkinson A, Kirkham JJ, Tudur-Smith C, et al. Reporting of harms data in RCTs: a systematic review of empirical assessments against the CONSORT harms extension. BMJ Open. 2013;3(9):e003436. 
14. Zorzela L, Golder S, Liu Y, et al. Quality of reporting in systematic reviews of adverse events: systematic review. BMJ. 2014;348:f7668.
15. Shinton R. Questions about authorisation of alteplase for ischaemic stroke. Lancet. 2014;384(9944):659-660.
16. Emberson J, Lees KR, Lyden P, et al. Effect of treatment delay, age, and stroke severity on the effects of intravenous thrombolysis with alteplase for acute ischaemic stroke: a meta-analysis of individual patient data from randomised trials. Lancet. 2014;384(9958):1929-1935.
17. Donaldson L, Fitzgerald E, Flower O, et al. Review article: Why is there still a debate regarding the safety and efficacy of intravenous thrombolysis in the management of presumed acute ischaemic stroke? A systematic review and meta-analysis. Emerg Med Australas. 2016;28(5):496-510.
18. Lenzer J. Alteplase for stroke: money and optimistic claims buttress the “brain attack” campaign. BMJ. 2002;324(7339):723-729.
19. Lundh A, Lexchin J, Mintzes B, et al. Industry sponsorship and research outcome. Cochrane Database Syst Rev. 2017;2:MR000033.
20. Hansen C, Lundh A, Rasmussen K, et al. Financial conflicts of interest in systematic reviews: associations with results, conclusions, and methodological quality. Cochrane Database Syst Rev. 2019;8(8):MR000047.
21. Institute of Medicine (US) Committee on Standards for Developing Trustworthy Clinical Practice Guidelines. Graham R, Mancher M, Miller Wolman D, et al, eds. Clinical Practice Guidelines We Can Trust. Washington, D.C.: National Academies Press (US); 2011.
22. Cain DM, Detsky AS. Everyone‘s a little bit biased (even physicians). JAMA. 2008;299(24):2893-2895.