Transcript Coaated stents: a new era

The DES controversy and DREAM
Eric J Topol MD
Professor of Genetics
Case Western Reserve University
Cleveland, OH
Robert M Califf MD
Professor of Medicine
Associate Vice Chancellor for Clinical Research
Director, Duke Clinical Research Institute
Duke University Medical Center
Durham, NC
Thumbs up / Thumbs down – Sept 22, 2006
Evolution of stent technology
Two problems with angioplasty technology
• Abrupt thrombosis at the time of the
procedure
• Restenosis
Better anticoagulation and the advent of
bare-metal stents helped resolve both
problems
Bare-metal stents reduced symptomatic
restenosis by 50% to 75%
Thumbs up / Thumbs down – Sept 22, 2006
Califf
Drug-eluting stents
Regulatory approval and marketing of drug-eluting
stents were based on short-term trials with a
straightforward population that excluded more difficult
PCI patients
The evidence was overwhelming
• No effect on risk of death or nonfatal MI in the 6- to
12-month timeframe
• Additional 70% reduction in restenosis
Two main types of drug-eluting stents
• Sirolimus-eluting (Cypher) stents
• Paclitaxel-eluting (Taxus) stents
Thumbs up / Thumbs down – Sept 22, 2006
Califf
Fantastic data . . . on the surface
The data looked fantastic
"Who wouldn't want to sign up for one?"
Renu Virmani (Cardiovascular Pathology, Gaithersburg,
MD) predicted that there would be a catastrophic
problem with the failure to cover the stent with
endothelium in a certain percentage of cases
Fairly early on there were publications of small series or
case reports of late thrombosis
• But the evidence seemed to indicate that it was not
much of a problem
Thumbs up / Thumbs down – Sept 22, 2006
Califf
PREMIER Registry
Funded by a company developing a drug to
treat angina
People who prematurely stopped clopidogrel
had a ninefold increase in the risk of stent
thrombosis
All acute-coronary-syndrome patients
• Much higher-risk patients than those
studied in clinical trials
Califf
Spertus JA, et al. Circulation 2006; 113(24):2803-2809.
Thumbs up / Thumbs down – Sept 22, 2006
BASKET-LATE
Swiss government–run trial
Results were "very concerning, almost
frightening"
After everyone had stopped taking
clopidogrel, there was a significant
increase in death and MI in patients
randomized to drug-eluting stents vs
those randomized to bare-metal stents
Califf
Pfisterer ME. American College of Cardiology Annual Scientific
Session 2006; March 14, 2006; Atlanta GA.
Thumbs up / Thumbs down – Sept 22, 2006
Signals from Barcelona
The possibility that there is a relatively linear but
consistent-over-time incremental risk with drug-eluting
stents compared with bare-metal stents
The additive effect over two to five years might be
highly significant, both clinically and statistically
None of the studies were definitive, but they all point in
the same direction
Dr Salim Yusuf concluded that if we stop putting stents
in people who don't need them, this would not be much
of a problem
Thumbs up / Thumbs down – Sept 22, 2006
Califf
SIRTAX
Examining rates of early and late stent
thrombosis
More than 8000 patients
By three years, the rate of stent
thrombosis with drug-eluting stents was
almost 3%
Topol
Wenaweser P. World Congress of Cardiology 2006;
September 3, 2006; Barcelona, Spain. Presentation 1012.
Thumbs up / Thumbs down – Sept 22, 2006
RAVEL: Five-year follow-up
Pristine from the standpoint of restenosis
More death with the Cypher stent than
with the bare-metal stent: 11.1% vs 7%
Not statistically significant, but
unexpected
Topol
RAVEL: Serruys PW. World Congress of Cardiology 2006;
September 6, 2006; Barcelona, Spain.
Thumbs up / Thumbs down – Sept 22, 2006
Meta-analysis
A significant excess of late thrombosis
with Cypher stents vs bare-metal stents:
6.3% vs 3.9%
Results are consistent with predictions
from BASKET
"I think it's a pretty significant concern"
Topol
Camenzind E. World Congress of Cardiology 2006;
September 3, 2006; Barcelona, Spain. Hotline I.
Thumbs up / Thumbs down – Sept 22, 2006
Courses of action
What advice would you give to a patient
who shows up in the cardiologist's office?
Thumbs up / Thumbs down – Sept 22, 2006
Califf
Not time to panic
Late thrombosis a very-low-frequency event
We don't know about the long-term safety of drugeluting stents in some individuals
We can't predict who is going to have a late clotting
risk
• Some patients never endothelialize the strut of the
stent because the drug is so effective
• Some patients have a propensity for thrombosis
Tell patients that there is a small risk over extended
follow-up
Thumbs up / Thumbs down – Sept 22, 2006
Topol
Restenosis: A trade-off
The trade-off: the low risk of stent clotting and
fatality vs the chance of having a repeat
procedure
Risk of a repeat procedure is higher with a baremetal stent
Restenosis is generally viewed as benign; this
might not be entirely true
We don't have good data from large, multiyear,
randomized trials comparing bare-metal and drugeluting stents
Thumbs up / Thumbs down – Sept 22, 2006
Topol
Genomics?
Weren't you arguing just last year that we
could do small trials with genomics and
not have to do large trials?
Thumbs up / Thumbs down – Sept 22, 2006
Califf
Genomics: The way of the future
If we had DNA for patients who had stent
thrombosis and for suitable controls, we
could identify the genes involved
Stent thrombosis doesn't appear to be a
mechanical issue (eg, stent apposition)
• Patients are either not endothelializing well
or are prone to thrombosis
Eventually, genomics will allow us to get a
handle on the problem, but we haven't even
started this work yet
Thumbs up / Thumbs down – Sept 22, 2006
Topol
Clopidogrel or not clopidogrel?
With drug-eluting stents, people are potentially
committed to dual antiplatelet therapy for the rest
of their lives
• The risk of bleeding
• The expense
• The possibility of having to stop clopidogrel
because of events such as significant dental
work, orthopedic procedures, other surgical
procedures
There are anecdotes of patients developing stent
thrombosis days after stopping dual antiplatelet
treatment
Thumbs up / Thumbs down – Sept 22, 2006
Topol
Practical advice
If you get a drug-eluting stent, stay on
clopidogrel until this is sorted out
If you have a patient you don't think can
stay on clopidogrel, try not to use a drugeluting stent
Thumbs up / Thumbs down – Sept 22, 2006
Califf
Weighing the risk
Bare-metal stents are preferable for the following
people:
• Those who don't take medication reliably
• Those who have issues with the cost
• Those who might require another significant
procedure
Until we have data from large, real-world,
multiyear trials, we must weigh the risk of
restenosis and its complications against the risk of
late thrombosis
Thumbs up / Thumbs down – Sept 22, 2006
Topol
FDA reform
The Future of Drug Safety: Promoting and
Protecting the Health of the Public
A report on postmarketing FDA issues
published by the Institute of Medicine
(IOM)
Thumbs up / Thumbs down – Sept 22, 2006
Califf
Hard-hitting report
A 255-page report that really slammed the
FDA
The recommendations are extraordinary
The IOM got paid $3 million to do it
• "I think the FDA would rather get its money
back."
It takes on problems with drug-safety
leadership
Thumbs up / Thumbs down – Sept 22, 2006
Topol
FDA takes action
The problem with drug-eluting stents is a
postmarketing issue
• There were grounds for their approval
The need for monitoring and proper
clinical trials is addressed in the report
The FDA is convening a special panel to
review this issue before year-end
Thumbs up / Thumbs down – Sept 22, 2006
Topol
The genie's out of the bottle
Drug-eluting stents have been on the market
for a number of years
They're used in 90% or more of stent
procedures in the US
Boston Scientific announced today that the
use of drug-eluting stents has diminished in
recent weeks
"It may be that the interventional community
is a little spooked about this late thrombosis
thing."
Thumbs up / Thumbs down – Sept 22, 2006
Topol
First responders
"I'd have to say the interventional
community is one of the most sensitive on
earth to clinical-trial findings and clinicalresearch statements"
In some other fields, it takes 15 years to
integrate results into clinical practice
Thumbs up / Thumbs down – Sept 22, 2006
Califf
News travels
There wasn't a very good showing of
interventional cardiologists from the US
at the meetings in Barcelona
• But it did seem to be the story that was
highlighted
Thumbs up / Thumbs down – Sept 22, 2006
Topol
Too many PCIs?
Salim Yusuf's research group (McMaster
University, Hamilton ON) continues to turn
out trials that are interesting and worthwhile
I don't think we are doing too many PCIs,
although we might be doing them in the
wrong people
"I think it's a procedure that has a lot of
benefit."
Thumbs up / Thumbs down – Sept 22, 2006
Califf
Attitudes in the US
In the interventional world, no one is
allowed to have ischemia, even if they're
asymptomatic
Yusuf believes that stable angina with
occasional angina or even a stressinduced functional test is acceptable
As long as there has been balloon
angioplasty, the attitude has been to
stamp out all ischemia
Is there a financial incentive to do this?
Thumbs up / Thumbs down – Sept 22, 2006
Topol
Financial incentive
As long as PCI is rewarded above other activities,
people are going to do it when it's a toss-up
In this country, the big money's in imaging, it's
not in PCI anymore
I don't think it's the only factor, but it is part of
the culture
When you hear there's a blockage and you can
open it up, it feels like a good thing
Thumbs up / Thumbs down – Sept 22, 2006
Califf
The debate continues
Drug-eluting stents became commonplace
because they got rid of the reoccurrence of
blockages
"It's a very blockage-centric world we live in in
the US."
There are no data to show that PCI in stable
angina benefits patients, but they do show that it
is beneficial in ischemia
Echo and nuclear imaging are going to be
emphasized with the CT angiography that's
popular now
Thumbs up / Thumbs down – Sept 22, 2006
Topol
Worth watching
The FDA hearing will be something worth
watching as it goes through these data
People should look at the FDA reform
report
Thumbs up / Thumbs down – Sept 22, 2006
Califf
The search for a cure for
"diabesity"
People are gaining weight and developing
diabetes at relatively young ages and
suffering the consequences thereof
The pharmaceutical industry has been
searching for drugs to prevent the
development of diabetes in people who
are at risk
Thumbs up / Thumbs down – Sept 22, 2006
Califf
DPP trial
The Diabetes Prevention Program (DPP) trial
• Looking at whether either diet and exercise
or the oral diabetes drug metformin
(Glucophage) prevent or delay the onset of
type 2 diabetes in people with impaired
glucose tolerance
Results
• An intensive exercise and diet regimen
reduces the time to onset of diabetes
diabetes by 60% to 70%
Thumbs up / Thumbs down – Sept 22, 2006
Califf
ACE inhibitors reduce diabetes
In populations studied for other reasons,
ACE inhibitors have been shown to reduce
the frequency of diabetes
These studies weren't looking specifically
at the diabetes end point in a rigorous
fashion
Thumbs up / Thumbs down – Sept 22, 2006
Califf
Another class of drug
Thiazolidinediones
• Affect the PPAR system (nuclear-receptor
hormones)
• Are widely used without any real outcome
data on their effect, other than lowering
blood sugar
Triglitazone showed a dramatic reduction in
time to onset of diabetes
• Unfortunately, it caused liver damage
Thumbs up / Thumbs down – Sept 22, 2006
Califf
The DREAM trial
Factorial randomized design
• 1/4 got rosiglitazone
• 1/4 got ramipril
• 1/4 got both rosiglitazone and ramipril
• 1/4 got placebo
Patients had impaired fasting glucose levels or
impaired glucose tolerance but no diabetes
The population was relatively young and at low
risk for cardiovascular events
DREAM was powered to show a reduction in the
onset of diabetes
Thumbs up / Thumbs down – Sept 22, 2006
Califf
DREAM: The ramipril arm
Reduction (about 1%) in onset of diabetes was
not statistically significant
Fasting glucose levels and the proportion of
patients who returned to normal glucose tolerance
significantly favored ramipril
No significant differences in clinical events
Absolutely no difference for the first couple of
years, then the curves began to separate but
didn't reach significance
It could be argued that this was a delayed effect
Califf
DREAM trial investigators. Effect of ramipril on the incidence of diabetes.
N Engl J Med 2006; 355.
Thumbs up / Thumbs down – Sept 22, 2006
DREAM: The rosiglitazone arm
Results with rosiglitazone
• A 60% reduction in the onset of diabetes
• Lower blood pressure
• A 2.4-kg increase in weight
• An increase in heart failure
• A trend toward an increase in vascular
events
Califf
DREAM trial investigators. Effect of rosiglitazone on the frequency of
diabetes In patients with impaired glucose tolerance or impaired
fasting glucose: A Randomised controlled trial. Lancet 2006.
Thumbs up / Thumbs down – Sept 22, 2006
A conundrum
ACE inhibitors work but have a much
smaller effect than expected
Thiazolidinediones clearly reduce
diabetes but have an uncertain negative
trend in terms of vascular events
Thumbs up / Thumbs down – Sept 22, 2006
Califf
Why didn't ramipril work?
The ramipril results were really surprising
Ramipril worked well in the HOPE trial
Other trials of ACE inhibitors showed a
pretty consistent reduction in the
incidence of diabetes
Of course, the DREAM patients didn't have
cardiovascular disease
Thumbs up / Thumbs down – Sept 22, 2006
Topol
DREAM population
The DREAM patients were younger than
the HOPE population and not as sick
Ramipril did work in the sense that
glucose levels were lower, it was just a
very weak effect
Thumbs up / Thumbs down – Sept 22, 2006
Califf
Very weak results
For primary outcomes over the 3.5 years,
the event curves (combined end point of
diabetes and death) are totally
superimposable
Are you impressed that there was a
consistent reduction in the incidence of
diabetes with ACE inhibitors in older
patients with cardiovascular disease?
Thumbs up / Thumbs down – Sept 22, 2006
Topol
More practical advice
"It's likely that that's true, but there's no
way to know for sure."
I wouldn't prescribe ramipril just to
prevent diabetes, but if I had a reason to
give an antihypertensive medicine to
someone in a prediabetic state, "it would
be a minor additional factor to make me
lean in that direction."
Thumbs up / Thumbs down – Sept 22, 2006
Califf
Surprising results
The fact that rosiglitazone was effective
for the combined end point (diabetes and
death) was a surprise
I knew that there would likely be some
increase in heart failure, as has been seen
with the other glitazones, but I didn't
expect it to be as marked
Thumbs up / Thumbs down – Sept 22, 2006
Topol
Drugs doing what they should
"I thought that this would work."
These drugs were designed to treat
diabetes, so it would be shocking if there
wasn't a reduction
Thumbs up / Thumbs down – Sept 22, 2006
Califf
No benefit
For the clinical outcomes, when mortality
was compared with the new incidence of
confirmed heart failure, there was no
benefit
Thumbs up / Thumbs down – Sept 22, 2006
Topol
The heart-failure issue
There is likely harm from the
thiazolidinediones that comes from fluid
retention and a proinflammatory state in
some patients
We've known about the heart-failure
issue for a long time
Thumbs up / Thumbs down – Sept 22, 2006
Califf
Why two papers?
Rosiglitazone might even be somewhat
worse than PYA or other agents in this
class
Why were these results not incorporated
into one paper?
Thumbs up / Thumbs down – Sept 22, 2006
Topol
Two for one
I presume that one of the advantages of a
factorial design is you get two trials for
the price of one
Thumbs up / Thumbs down – Sept 22, 2006
Califf
Two thumbs up
This trial asked a legitimate question
"The results are disappointing in that it doesn't
really change my view of prescribing ramipril."
• It's a great drug for hypertension and LV
dysfunction
Ramipril probably somewhat reduced coronarydisease events, but I wouldn't prescribe it just to
prevent diabetes
Rosiglitazone does show that if you have a drug
that's highly effective, you might make a dent in
the long-term issue of diabetes
Thumbs up / Thumbs down – Sept 22, 2006
Califf
Is it worth the trade-off?
The adverse effects weren't significant,
but they weren't far from being
significant
By inhibiting the development of diabetes,
you invite other adverse outcomes
Thumbs up / Thumbs down – Sept 22, 2006
Topol
Needed data on its way
Until we have more outcome data, we should
not be prescribing thiazolidinediones to
prevent diabetes
The good news is that there are several major
cardiovascular trials under way that will
provide answers
Thumbs up / Thumbs down – Sept 22, 2006
Califf
Another two thumbs up
Aside from the concern about the reporting,
it's a two-thumbs-up trial
The prevention side of rosiglitazone is
something new
The adverse outcomes are, at the very least,
disquieting
Cardiologists don't get to see this population
until a decade later
Thumbs up / Thumbs down – Sept 22, 2006
Topol