Transcript Topol - Medscape

Carvedilol or Metoprolol European
Trial (COMET)
Eric J Topol MD
Provost and Chief Academic Officer
Chairman, Department of Cardiovascular Medicine
The Cleveland Clinic Foundation
Cleveland, Ohio
Robert M Califf MD
Professor of Medicine
Associate Vice Chancellor for Clinical Research
Director, Duke Clinical Research Institute
Duke University Medical Center
Durham, North Carolina
Thumbs – July 2003
Enrollment
COMET enrolled 3029 patients with class
II to IV heart failure.
• 1511 patients assigned carvedilol (target
dose 25 mg twice daily)
• 1518 patients assigned metoprolol
(target dose 50 mg twice daily)
Primary endpoint: All-cause mortality and
composite of all-cause mortality and allcause hospitalization
Thumbs – July 2003
Trial history
Recruitment started December 1996,
ended January 1999
Trial was focused on which beta-blocker
was better -- a challenge to the
uniformity of a class effect
Thumbs – July 2003
Primary results
Carvedilol
(n=1511)
(%)
Metoprolol
(n=1518)
(%)
p
All-cause
mortality
33.9%
39.5%
0.0017
All-cause
mortality
or all-cause
hospitalization
73.9%
76.4%
0.1222
Endpoint
Thumbs – July 2003
Poole-Wilson PA et al. Lancet 2003;362:7-13
Short-acting metoprolol
Concerns raised over the dose and
formulation of metoprolol used
• Short-acting metoprolol was used
instead of the long-acting metoprolol
which is the current standard
Thumbs – July 2003
Class effect
“My feeling about the study overall is
that it is important because it does
challenge the class effect.”
Califf
• All beta-blockers do not reduce
mortality the same
• Metoprolol not the same as
formulation used in MERIT-HF
Thumbs – July 2003
HR reductions and beta blockade
HR rate decreased more with
carvedilol initially but no differences
reported over time
HR reductions in the first few months:
• 13.3 beats per min with carvedilol
• 11.7 beats per min with metoprolol
• After 16 months, no differences in HR
Thumbs – July 2003
Which comparison?
“At the end of the day we’re left with
the question, ‘Is this comparing
different levels of beta blockade or is
it comparing two different beta
blockers with different biological
properties that produce different
outcomes?’”
Califf
Thumbs – July 2003
Perplexing
“I’m fairly perplexed about COMET.”
Topol
• Carvedilol may have distinct biological
properties
• Dosing of beta blocker remains an
issue, as reflected in HR analysis
Thumbs – July 2003
Loaded trial?
“While there are some biologic basis
where these two drugs from the beta
blocker class could be differentiated,
I’m not sure the trial wasn’t
engineered, or loaded, to
demonstrate carvedilol’s
superiority.”
Topol
Thumbs – July 2003
Valid comparison?
There was not equally effective betablockade
Carvedilol appears more effective at
the moment but we aren’t sure of
the validity of the comparison
Topol
Thumbs – July 2003
Questions remain
The positioning of carvedilol as uniquely
superior may not be appropriate
Perplexing questions remain because of
the dosing choices and the short acting
vs long acting forms of the drug
Topol
Thumbs – July 2003
Was it a fair match?
“At the moment, you’d have to
conclude, based on the evidence,
that carvedilol is the winner, but was
it a fair match?”
Topol
Thumbs – July 2003
Background therapy
Diuretic
ACE inhibitor
Digitoxin
Warfarin
Aspirin
Spironolactone
ARB
Thumbs – July 2003
COMET (n=3029)
99%
91%
59%
46%
37%
11%
7%
Issue of dose
“All COMET is showing is that a higherdose regimen is better than a lower
one. Not very exciting.”
Dr Ake Hjalmarson, principal
MERIT-HF investigator
“I am saying that it [carvedilol] is better
than metoprolol tartrate 50 mg twice
daily, and this must be due to actions
other than beta-1 blockade.”
Dr Milton Packer
Thumbs – July 2003
Treatment options
What’s the right way to go about
deciding about which treatment is
best?
The comparative analysis once a
product is on the market is often
controlled by industry
Califf
Thumbs – July 2003
Neutral court
Hard to believe COMET investigators
would proceed with a clinical trial
skewed to favor carvedilol over
metoprolol
Califf
Need for neutral court to arbitrate “fair
and square” comparative analysis
between active controls
Topol
Thumbs – July 2003
Consumer Reports
Consumer Reports exists for consumer
products, and is trusted to provide
independent evaluations of products
“In the arena that’s most directly related
to our quality of life and longevity, we
leave it up to companies which are
making the products to do their own
comparisons, to control the data and
very often control the publication.”
Califf
Thumbs – July 2003
Not happy
“I don’t think [COMET] was rigged or
engineered but I’m not happy with it.”
Califf
If trials were designed in an open system
COMET may have been designed
differently
Thumbs – July 2003
A case of cynical MDs?
“The sponsor is GlaxoSmithKline and
[COMET] comes out in their favor.
That’s at least a concern in the cynical
world that we live in. Was there
something about the trial that wasn’t
completely level?”
Topol
Thumbs – July 2003
Bold trial
Getting a company to do an active control
trial against another product, it’s very
hard.
Califf
Usually you would expect that if you’re
going to take on a drug in the class,
you’re looking at non-inferiority
designs.
Topol
Thumbs – July 2003
Issues in trial design
The label doses of the drugs were used,
which is reasonable but leaves the
question open as to whether the dosing
is correct
The science has gotten far ahead of our
ability to get the trials done in two ways
• Size of the study
• Speed of making the comparisons
Modest differences become very
important
Thumbs – July 2003
Califf
Conclusions from COMET
“The results are supportive that, until
proven otherwise, carvedilol is the
winner; is the drug of choice.”
Topol
• Carvedilol extends survival compared
to metoprolol 50 mg twice daily
Thumbs – July 2003
Carvedilol at the CCF
“Our heart failure people are not thoroughly
convinced…but it’s a very large group here
and there is some inconsistency.”
Expense on a long-term basis is an important
factor
“It isn’t a definitive knock-out of metoprolol
XL here.”
Topol
Thumbs – July 2003
Carvedilol at Duke
Patients who want to know everything and
can afford it are mostly picking carvedilol
“The majority of people [make a decision]
based on some gestalt about what their
pocket book can tolerate and what kind of
insurance they have.”
“While the nod goes to carvedilol, a lot of
people are getting prescribed metoprolol
because of the combination of the
uncertainty and the cost.”
Califf
Thumbs – July 2003
COMET vs GUSTO I
It is surprising Milton Packer made the
comparison with GUSTO I (tPA vs SK)
Indisputable survival difference
Dose of streptokinase used was the one
still used today while in COMET the
dose issue is at the heart of the
controversy
Topol
Thumbs – July 2003
Deficiencies in design
Payers are left with difficult, multimillion
dollar decisions with a lack of decisive
data
“It’s just not a good national strategy to
leave this up to sales reps to convince
the doctors of what to prescribe.”
Califf
Thumbs – July 2003
Changing infrastructure
NIH feels we need to re-invigorate
research infrastructure to get these
questions answered
“It seems the people at the end of the line
of the research are having the least say
in how the money is spent on research
and I think that will change.”
Califf
Thumbs – July 2003
COMET trial review
Dr Eric Topol
1 thumb up
“I don’t think it measures
up to the optimal clinical
trial.”
Advances the field with
some confusing results
Thumbs – July 2003
COMET trial review
Dr Rob Califf
1 thumb up
“It does clarify the
questions but leaves us
with an answer which is
highly debatable.”
Thumbs – July 2003
Carvedilol or Metoprolol European
Trial (COMET)
Eric J Topol MD
Provost and Chief Academic Officer
Chairman, Department of Cardiovascular Medicine
The Cleveland Clinic Foundation
Cleveland, Ohio
Robert M Califf MD
Professor of Medicine
Associate Vice Chancellor for Clinical Research
Director, Duke Clinical Research Institute
Duke University Medical Center
Durham, North Carolina
Thumbs – July 2003