Title: Clinical Trial Commentary
1Clinical Trial Commentary
GUSTO V
- Dr Eric Topol
- Provost and Chief Academic Officer
- Chairman and Professor, Department of Cardiology
- Cleveland Clinic
- Dr Robert Califf
- Professor of Cardiology
- Associate Vice Chancellor for
- Clinical Research at Duke University
2Study design
GUSTO V
Randomization
Standard-Dose Reteplase (10 10 U Double Bolus)
Abciximab Low-Dose Reteplase (5 5 U Double
Bolus)
Heparin 5000 U 1000 U/hr (800 U/hr for lt70 kg)
Heparin 60 U/kg (max 5000 U) 7 U/kg-hr
3Endpoints
GUSTO V
- Primary
- mortality (all-cause) by 30 days
- Secondary
- mortality (30-day) or non-fatal disabling stroke
(in-hospital or 7-day) - hemorrhagic stroke (in-hospital or 7-day)
- mortality by 1 year
- reinfarction
- coronary revascularization
- other prespecified complications of MI
4Statistical methods
GUSTO V
Superiority Testing one-sided Type I error lt
2.5 for control mortality rates ranging from 5 -
9. approximately 80 power to detect 15
reduction if control mortality rate
7.4 Non-Inferiority Testing less than 10
relative increase in mortality - upper bound of
95 CI for relative risk 1.10 one-sided Type I
error ranges from 2.051 - 2.627 for control
mortality rates ranging from 5 - 9
5Primary endpoint
GUSTO V
Non-inferiority boundary
Upper bound of 95 confidence interval 1.076
6Reinfarction
GUSTO V
7Revascularization
GUSTO V
8Non-fatal complications
GUSTO V
- It looked like, if one starts to consider the
whole gestalt of non-fatal complications, that
there was a very consistent and important
reduction of these endpoints for the
combination. -
- Dr Eric Topol
- Provost and Chief Academic Officer
- Chairman and Professor, Department of Cardiology
- Cleveland Clinic
9Bleeding
GUSTO V
Abciximab Reteplase Reteplase N 8260
N 8328 EENT () 0.1 0.6 Pulmonary
() 0.1 0.3 Cardiac () 0.1 0.1 Retroperitoneal
() 0.1 0.1 Genitourinary () 0.1 0.4 Sheath Site
() 0.7 0.4 Gastrointestinal () 0.4 1.9 Other
Puncture Site () 0.3 0.6 Surgical () 0.4 0.3
10Doubts on non-inferiority
GUSTO V
- Accusation We just cooked up this
non-inferiority thing, mortality reduction is all
that counts. - Califf
- Rebuttal The overall mortality was extremely
low, and the improvement in the combination arm
was flanked by other improvements. - Topol
11Beyond 30 days
GUSTO V
- We have to start getting beyond just life or
death at 30 days. The SHOCK trial taught us
a big lesson, that you don't always see the
benefit of an aggressive strategy for cardiogenic
shock at 30 days, in fact you see a lot more
impact of this at 1 year. I think we may well
see the same thing as far as 1 year mortality in
GUSTO V. -
- Dr Eric Topol
- Provost and Chief Academic Officer
- Chairman and Professor, Department of Cardiology
- Cleveland Clinic
12An entirely new strategy
GUSTO V
- We did do what we had hypothesized we could do.
Which is develop an entirely new strategy, not
one that was red clot dissolving, to achieve a
very impressive endpoint of mortality at 30 days,
and beyond that. -
- Dr Eric Topol
- Provost and Chief Academic Officer
- Chairman and Professor, Department of Cardiology
- Cleveland Clinic
13Mortality results are biased?
GUSTO V
- Accusation The smart doctors just siphoned off
the high-risk patients for direct angioplasty. - Califf
- Rebuttal Many of the patients were outside the
US, where cath-based reperfusion isn't the
standard mode. But there doesn't seem to be a
tendency towards low-risk patients in the trial. - Topol
14Final enrollment
GUSTO V
15The wrong lytic?
GUSTO V
- Accusation Reteplase is a weak lytic and was a
bad choice for the trial. - Califf
- Rebuttal We have no head-to-head comparative
data. Without the head-to-head it's too much
speculation. - Topol
16Non-fatal MI questions
GUSTO V
- Accusation The non-fatal MI wasnt strictly
defined and isn't useful. How can you have a big
difference in MI but not mortality? - Califf
- Rebuttal After mortality, death of heart tissue
is the most important thing. These were major
clinical events linked to other complications
seen in the trial. - That it was only day 7 and non-blinded data are
legitimate critiques. - Topol
17CURE trial comparison
GUSTO V
GUSTO 5
CURE
Reteplase Aspirin Reteplase Abciximab RR
Aspirin Clopidogrel RR Death 5.9 5.6 0.95 5.5 5.1
0.92 MI 3.5 2.3 0.67 6.7 5.2 0.77 Stroke 0.3 0.2
0.76 1.4 1.2 0.85 Transfusion gt
2U 3.7 5.0 1.38 2.2 2.8 1.28
18Importance of reinfarction
GUSTO V
- GUSTO I and III showed a marked difference in 1
year survival for those who had no reinfarction
in 30 days vs those who did. - More reason to suspect we should see an even
stronger difference in mortality at 1 year. - Topol
19Transfusions
GUSTO V
- "But the question is death of heart tissue or
death of patient vs a transfusion. When you look
at the net there that maybe you're better off
reducing the death of the patient or the death of
heart tissue and you have to bite the bullet with
transfusion. -
- Dr Eric Topol
- Provost and Chief Academic Officer
- Chairman and Professor, Department of Cardiology
- Cleveland Clinic
20Bleeding
GUSTO V
- Bleeding is clustered in the elderly, female, and
light-weight patients. - Different anti-coagulants may lower this bleeding
even further. - Topol
21Problems with the trial
GUSTO V
- The lack of mortality reduction was
disappointing. - GUSTO I reduced mortality by gt 14 and some still
said we didn't reduce mortality. - There are always nay-sayers for any large trial.
- Topol
22Time will tell
GUSTO V
- "The only way to know what you've done, is
how the trial's data are adopted in practice. -
- Dr Eric Topol
- Provost and Chief Academic Officer
- Chairman and Professor, Department of Cardiology
- Cleveland Clinic
23Embracing the results
GUSTO V
- The costs of the combination therapy should not
be very different from the standard so that isn't
fueling the controversy. - I would think it should be viewed as a good
thing - reduced non-fatal endpoints
- discriminates the population at risk of
bleeding - Bleeding didn't override the clinical benefits
- This should be embraced for certain patients.
- Topol
24Apply it to practice?
GUSTO V
- I'd like to see any better data on how to treat
patients today. - There's a cath-lab strategy, but often there is a
delay, and most places don't have it available. - It may not be for all patients. (Tough to
advocate for patients with small MIs) - Topol
25Cooking up the cocktail.
GUSTO V
- Reteplase currently comes in two vials. So you
use just one with the abciximab. - Costs about 300 more than reteplase or
tenecteplase alone. - There are several hospitals that have done it for
the last year, even withou the GUSTO V data. - Topol
26Who to treat
GUSTO V
- Patients with significant MIs
- Patients 75 years old or younger
- If it is a relatively small MI, I probably would
NOT bother using combination therapy. - Topol
27ASSENT III
GUSTO V
- Assent III should offer some supporting evidence.
Not as large a trial, but it should shed further
light on the question. - Califf
28Faster treatment
GUSTO V
- The 90 minute to 2 hour delay getting to cath lab
is the big question. Would we be better off
having drugs working en route? - Topol
- The great hope is that we can organize things to
treat people quickly and open the artery and the
cath-lab is proving where you want to be in the
long run. - Califf
29Reservations
GUSTO V
- I think it's a matter of getting organized and
absorbing the data some more and seeing whether
ASSENT 3 confirms it. I think it is so close
temporally that I'm not quite ready to jump on it
at this point." - Dr Robert Califf
- Professor of Cardiology
- Associate Vice Chancellor for
- Clinical Research at Duke University
30Other combinations
GUSTO V
- All combinations are possible, but you can't
adopt any combination until you have some solid
evidence with a large-scale trial. - GUSTO V is favorable on balance, but it is
tenuous, a small difference. - Strong data is needed on other combinations
before we can advocate them. - Califf
31Intracranial hemorrhage
GUSTO V
Abciximab Reteplase
Reteplase
Odds Ratio 95 CI
Intracranialhemorrhage rate
0.6
0.6
Age
0.045
lt 45
0.2
0.1
0.021
gt 45 - 55
0.3
0.1
gt 55 - 65
0.4
0.4
gt 65 - 75
1.0
0.8
1.1
2.1
gt 75
0.1
1
10
ReteplaseBetter
Abciximab Reteplase Better
32Lack of progress on ICH
GUSTO V
- Trial didn't show any increase in ICH overall.
But it remains a problem with the elderly. It
doesn't look like a great strategy for the
elderly. - Topol
- Most frustrating to me is that we have made no
progress on ICH. We still don't know how to pick
out people at risk. - Califf
33Compared to other trials
GUSTO V
ICH Rates
N 18,495 15,059 16,949 15,078 16,588
34The naysayers
GUSTO V
- "The most frustrating thing is to see that no
matter what trial you do, no matter what the
findings are, they are very harshly criticized by
some. And after a while it makes you not want to
be engaged in clinical trials. " -
- Dr Eric Topol
- Provost and Chief Academic Officer
- Chairman and Professor, Department of Cardiology
- Cleveland Clinic
35Stepwise progress
GUSTO V
- We need to remember that AMI is still the
developed world's number 1 cause of death and
disability. - Anything we do to chip away at the problem is a
step-wise advance. - Huge reductions in mortality arent always
possible. - Topol
36Fast track publication
GUSTO V
-
- With the agents already available, getting the
information out to the medical community quickly
and accurately was important. - Topol
-
- Making sure things get published before all the
rumors start flying around is a laudable goal. - Califf
37GUSTO V trial review
GUSTO V
- Dr Eric Topol
- Two thumbs up
- I'm not saying that's what the findings
necessarily support but I think in terms of the
design."
38Importance of non-inferiority
GUSTO V
- We want to have therapies that have fewer side
effects, or are easier to give, or cheaper.
Hopefully a combination of all of those. In many
cases you may not have a reduction in mortality
but you sure want to make sure that you dont
create an excess mortality." - Dr Robert Califf
- Professor of Cardiology
- Associate Vice Chancellor for
- Clinical Research at Duke University
39GUSTO V trial review
GUSTO V
- Dr Robert Califf
- Two thumbs up
- "A somewhat biased two thumbs up on both
accounts."