July 2005
Feature Story
A cardiac marker test that is FDA-approved, included in the Medicare
laboratory fee schedule, and proven to ease some of the pressure of high-stakes
triage would seem to have the makings of a clinical mainstay. And a diagnostic
test for ischemia-modified albumin, or IMA, is in fact steadily making
inroads into routine emergency room laboratory testing of chest pain patients.
But until recently, the test has kept a low profile. While almost every
emergency department does cardiac marker testing, "If you ask the average
ER physician about IMA, they’d probably say they don’t use it," says Alan
Wu, PhD, chief of clinical chemistry at San Francisco General Hospital
and professor of laboratory medicine at the University of California.
About 6 million people who go to the ER every year because of chest pain
are admitted for evaluation, but at least a million of them won’t end
up diagnosed with an acute coronary syndrome, or ACS. A $40 to $50 blood
test that makes a large number of those hospital overnight stays and stress
tests unnecessary would seem to be a shoo-in.
The albumin-cobalt binding assay, used to measure ischemia-modified albumin,
was approved by the Food and Drug Administration because it does just
that.
IMA test developer Ischemia Technologies (acquired in February by Inverness
Medical Innovations) established the test’s value in ruling out a future
acute ischemic event&—meaning that ER physicians can make some "go/no-go"
decisions to discharge low-risk patients when they have a nondiagnostic
ECG and negative results on both IMA and troponin.
Recently, even more evidence has emerged to support this "rule-out" function
of the marker. W. Frank Peacock, MD, emergency medicine vice chief for
research and medical director of event medicine at the Cleveland Clinic
Foundation, presented a meta-analysis of several studies in April.
"We looked at more than 1,000 people who had an IMA done when they arrived
at the ER with a reasonable chest pain story. They had a negative troponin
and nondiagnostic EKG, and we looked at the sensitivity of IMA for an
acute event, meaning angioplasty, death, or a cardiac ischemic event."
The analysis found that the predictive value of a negative IMA test result
(for the absence of angioplasty, death, or a cardiac ischemic event) is
over 95 percent, Dr. Peacock told attendees at a meeting of the Society
for Academic Emergency Medicine.
After looking at some studies of outcomes over 30 days, he found that
if the IMA test was negative the patient did very well, though for six-month
outcomes it was not as good a predictor. "But neither was stress testing,"
Dr. Peacock says.
Traditional stress testing, as well as myocardial perfusion imaging,
when used to rule out acute ischemic events within six months after hospitalization
have no statistical predictive edge over IMA&—yet the cost of an
average visit can exceed $4,500.
That makes an IMA test look enticing indeed. "If I can do the same thing
with equal accuracy with a blood test, and it saves patient admissions
overnight, that’s a real opportunity," Dr. Peacock says.
"The number of people who present with chest pain and stay at the hospital
today is close to 80 or 90 percent. Because of the concern about bad outcomes,
most people who go to the ER get admitted either to the hospital or to
a chest pain evaluation unit, except for the clearly obvious noncardiac
cases.
"Other than those, anyone with a remote possibility stays a long time—at
least many hours—while they do cardiac marker testing and possibly
stress testing," Dr. Peacock says.
That amounts to millions of admissions each year. "It’s a huge number,
and any safe reduction in the number of people we currently admit is an
improvement."
But what’s holding up wider use of the IMA test, according to Ischemia
Technologies, is a shortage of instruments that can perform it. "The thing
that has precluded our broad growth has been instrument platforms," says
Donna Edmonds, senior vice president of Wampole Laboratories, distributor
of the IMA test and a subsidiary of Inverness.
The IMA test is currently available on the Roche/Hitachi 917/Modular
P, Roche/Hitachi 911, Roche Cobas Mira Plus, and open channel of the Beckman
Coulter Synchron LX 20.
But this summer, Edmonds says, Inverness plans to launch the test on
Roche’s Integra 700/800 instrument series, which will make the IMA test
available on essentially all Roche analyzers, both open- and closed-channel
systems.
"That launch would give Roche a unique position in the market," she says,
but discussions with the other major chemistry analyzer vendors are underway
for additional instrument platforms. In 2006, the company also expects
to add the IMA test to an acute coronary syndrome platform for point-of-care
testing.
As a venture capital-backed company, Ischemia Technologies lacked the
marketing clout that Inverness brings to the table. Now IMA is the first
release product under the umbrella of Inverness Medical Cardiology Systems,
Edmonds says.
"Inverness made an extremely strategic investment in getting the test
for the albumin marker and it is investing heavily in exploiting the IP
it purchased with IMA, as well as other markers that will be complementary
to IMA."
But some researchers believe the jury is still out on how IMA works and
what clinical predictive ability it offers. A study published in the June
21 issue of the Canadian Medical Association Journal casts doubt
on IMA as a marker of cardiac ischemia (Worster A, et al. Ability
of ischemia-modified albumin to predict serious cardiac outcomes in the
short term among patients with potential acute coronary syndrome.
2005;172[13]:1685-1692).
It found that in 189 patients presenting to an emergency department within
six hours after chest pain, IMA was a poor predictor of serious cardiac
outcomes in the short term.
"These results attest to the necessity for thorough evaluation of new
diagnostic tests before their dissemination," the study authors concluded.
An important limitation of IMA testing is the short window in which IMA
actually rises. "With an ischemic event, IMA rises very rapidly, but it
clears very rapidly too. If patients haven’t had chest pain for a while,
then the IMA is going to be negative," Dr. Peacock explains.
"The patient must have had ischemia occur within three hours of arrival
at the ER. Otherwise the IMA test may be a false-negative."
Some experts believe this shouldn’t be a problem. "You’re not going to
be doing testing after six hours," Dr. Wu points out, "and the ischemia
window actually complements the window for troponin, because it doesn’t
become positive for three to six hours, and once troponin is positive
you will not need an IMA test."
However, the CMAJ study gives another reason for caution, says
Marc S. Sabatine, MD, MPH, investigator with the TIMI (Thrombolysis in
Myocardial Infarction) Study Group and associate physician in the Division
of Cardiovascular Medicine, Brigham and Women’s Hospital, Boston.
"Although the negative predictive value of IMA is reasonable, as so many
patients are IMA-positive, the ability to rule-out ACS would apply to
only a small minority of patients."
That’s one reason his hospital does not use IMA. He says, "I think more
research is necessary before we can state with confidence the role IMA
should play in diagnosis and prognosis in patients with chest pain."
Fred Apple, PhD, medical director of clinical laboratories for Hennepin
County Medical Center and professor of laboratory medicine and pathology
at the University of Minnesota in Twin Cities, is another who has opted
against IMA use in his laboratory. In his view, the FDA approved the test
inappropriately following studies that had too few subjects to validate
that the test works.
Dr. Apple’s laboratory served as a core laboratory for Ischemia Technologies’
research on IMA as well as other studies. One study published monitoring
IMA and troponin levels in marathon runners and suggested that IMA increases
were likely to be not from the heart but from another source, such as
from gastrointestinal ischemia. The impact of interferences causing falsely
low IMA levels due to lactate and free fatty acids was not described.
"I don’t think IMA is actually a test specific for myocardial ischemia,"
he says. "Investigators have been studying this marker for five to seven
years, and there’s still no definitive paper that’s convinced me this
will add value."
"I know if it is negative, the test appears helpful, but I live in a
world where clinicians want to know what the positive test means also,
and as the company acknowledges this test has false-positives, I can’t
bring it into my laboratory under these conditions."
"Among basic scientists rather than clinicians, I think there’s skepticism
because of the lack of proof of what this molecule IMA really is," Dr.
Apple adds. "It’s not been proven what is going on in the physiology of
this molecule we’re supposed to be measuring."
"For example, studies show a patient having very high levels-then 10
minutes later it’s gone. Scientists want to know why an albumin molecule
with a half-life of 10 days disappears so quickly."
The main public health issue in cardiology is that there are high-risk
patients with ischemia who are now undetected, says Rob Christenson, PhD,
professor of pathology at the University of Maryland School of Medicine
and director of rapid-response and point-of-care laboratories, University
of Maryland Medical Center.
"That would be the real target: To detect those patients and get them
into a treatment plan to mitigate that risk," says Dr. Christenson, who
is a coauthor with Dr. Peacock on the meta-analysis.
"But the second issue is cost. Many medical centers now just admit patients
with suspected cardiac ischemia, so anybody you could release would bring
an economic benefit." If the test results are true negatives in even a
minority of patients, then the test would still be useful, because "that’s
a group that would have been admitted before."
"We have excellent markers of necrosis, the cornerstone of which is troponin,
but so many patients even though they have a negative troponin have relatively
high mortality—or other adverse outcomes," Dr. Christenson points
out.
"So we’re only doing part of the job. There is clearly a need to detect
patients who are in the part of the ACS continuum where they have reversible
cell injury and no cell death leading to troponin release, even with later
temporal measurement." These patients still have unstable plaque and thrombus
formation, he says, which means they’re at high risk of a subsequent cardiac
event, which could be myocardial infarction or sudden death.
According to anecdotes from cardiologists, Dr. Christenson says, "unstable
angina is probably one of the most challenging clinical diagnoses there
is for specialists." One can imagine, then, what it’s like for ER physicians,
he says, who aren’t necessarily specialists. "So an objective marker would
be a useful additional indicator. That’s how I got involved in it."
Dr. Christenson describes risk stratification as a "stoplight" phenomenon.
"There are a lot of patients who have elevated IMA values. The question
is, what do they mean? The challenge is to establish guidelines based
on evidence. For example, you might have ’green,’ which is below a safe
cutoff and represents true negatives; ’amber,’ which means equivocal and
confers uncertainty; and ’red,’ which means there is a high probability
this is a high-risk patient."
Right now no other "rule-out" test has been approved by the FDA, but
a dozen other biomarkers are in various stages of development, and they
will compete with IMA as an early marker for rule-out and risk stratification,
Dr. Wu explains.
He is optimistic that IMA will prove useful as a "rule-in" test for ischemia
as well. "Troponin only tells you a heart attack has occurred," he says.
Because troponin predicts a likely future event, it’s also useful for
risk stratification, "but when you see troponin you have evidence of myocardial
damage. You may never get a positive troponin in a patient with ischemia
but you will with the IMA test."
"That tells you a heart attack is likely to occur in the very near future."
He contrasts IMA with cholesterol testing. "Cholesterol can tell you you’ll
have a heart attack sometime in the next 20 years. Ischemia testing says
it’s imminent—and that’s the exciting part of this story."
Dr. Peacock, who works three days a week in clinical care, says IMA-positive
results can be provocative. "Sometimes you get high numbers and say, ’holy
smokes, what’s that?’ There is some data to suggest high numbers like
110 on an IMA are significant, and it’s getting studied."
"But it’s not been answered. We don’t have much guidance on appropriate
action to take. Should you do an intervention on a patient? We just don’t
know."
"The party line is if you had somebody with a planned course of actions,
admission to a telemetry bed or observation unit, and the IMA is positive,
do what you were planning to do. If the IMA comes back negative, you may
be able to downgrade your plan based on a negative result. But with a
positive, no one is quite sure what to do," Dr. Peacock says.
Dr. Wu, a member of the scientific advisory board for Ischemia Technologies,
says neither San Francisco General Hospital nor his previous employer,
Hartford (Conn.) Hospital, offers IMA.
"It’s well known that the test suffers from lack of specificity," Dr.
Wu explains. "It’s pretty clear what a negative result means, but unfortunately
people want to know what a positive result means. A negative result is
helpful enough because it’s useful in ruling out ACS in the ER, but apparently
that’s not been enough for a lot of laboratories to justify offering the
test."
The decision makers would be the laboratory and the ER, "and in this
case, cardiology," Dr. Wu adds. "First of all the test is not offered
on very many platforms or by a point-of-care testing device. Right now,
it’s only really on Beckman and Roche analyzers, so if you don’t have
any of those instruments, the laboratory will not be able to provide this
service."
He agrees that it’s also not completely clear what the test is. "The
mechanism of how albumin is modified and why is not completely elucidated.
People have to pretty much accept it at face value if they can’t explain
it."
"Research has been going on since the beginning, and different theories
have been put forth and they keep changing, so that’s a source of some
concern—maybe more on the part of laboratory people because the
ER wouldn’t care as much," says Dr. Wu.
Finally, awareness about the test is not as high as it should be. "It’s
difficult to reach out to everybody in terms of public information," he
says.
The new meta-analysis, however, which is under review at a cardiology
journal, should help. "The review paper found that positive results indicate
risk stratification for adverse events, especially if the test is used
in conjunction with an electrocardiogram and troponin. When it’s published,"
Dr. Wu says, "the paper will help tremendously in getting this test more
accepted."
The University of Maryland doesn’t have the platforms on which the IMA
test is now being run in the routine lab. "We have used the test, but
we’re not using it routinely," Dr. Christenson notes, adding that the
medical center has a Roche Cobas Mira in the research laboratory only.
Adaptation to an array of platforms is what’s limiting the test, he says.
"You have the choice of making it available on many different platforms,
or more like the Biosite BNP test where you have one platform that is
small and inexpensive, and anybody can have access and get it implemented."
But now that Inverness Medical Innovations has acquired Ischemia Technologies,
he expects much broader availability.
Raising clinician awareness is the other key plank of Inverness’ strategy
for IMA, Donna Edmonds says. Since the acquisition, "we haven’t done any
advertising because we needed additional platforms."
With the Cobas-Integra 700/800 launch and the ability to be on all Roche
platforms, the company plans to advertise and hopes to gain visibility
when the results of Dr. Peacock and Dr. Christenson’s study are published.
More ambitiously, investigators from Massachusetts General Hospital are
finalizing a protocol for a 1,250-patient study called IMAGINE, which
will use five sites in the U.S. and one or two in Europe to establish
the utility of IMA.
Principal investigator James Louis Januzzi, MD, co-director of the coronary
care unit at Massachusetts General and assistant professor of medicine
at Harvard Medical School, hopes the study will launch later in the summer
and supply the missing link in research to date.
"Most of the studies so far have been small, single-center studies. What
most people are interested in is a large-scale prospective study looking
at the utility of the marker in evaluating patients with chest pain."
The goal of the IMAGINE study will be to confirm that IMA has good negative
predictive value for excluding patients at low risk, and to evaluate "what
a very high IMA value in a chest pain patient truly means," Dr. Januzzi
says.
"Can we harness the positive predictive value of the marker in patients
thought to be ischemic? Will it tell us something bad is going to happen
even before other markers like troponin?"
IMAGINE will look at multi-ethnic and multinational data to give a real-world
assessment of how IMA works in patient populations not addressed in prior
studies, he says, "so we can say beyond any shadow of a doubt that we
can apply the marker to emergency room patients."
A key complication, Dr. Januzzi notes, is that the potential utility
of the marker changes with respect to the type of physician use. "ER physicians
are looking for a rule-out, so IMA’s high predictive value makes the marker
a winner because they’re turning to the marker for that strength."
"In the cardiology world," he says, "we’re looking for markers to tell
us about patient risk, much the way troponin and BNP do. It’s important
for me as a cardiologist to validate the need for that kind of marker.
That’s why we’re exploring what IMA’s utility might be."
"My suspicion is there is probably a very high threshold, probably up
around 120 units, above which the likelihood of cardiovascular complications
in a patient with chest pain rises considerably."
IMA’s potential benefits will be best determined from a large study,
"which will allow us to look at the data critically and look at cut points
very carefully," Dr. Januzzi adds.
Dr. Wu suggests that other studies will have to be conducted to show
that when used properly, the IMA test reduces ER visits and costs.
"It’s intuitive that that should happen, but it’s also incumbent on ER
physicians to take a negative value and actually rule out ACS and send
the patient home. And there may be some hesitancy to do that based on
a laboratory test."
"There’s no question about the need for a marker of ischemia and a marker
for plaque instability," Dr. Wu says. "That’s indisputable. And that’s
why there’s so much research in that area. The skepticism arises from
deciding which test to use. Will there be one test or a panel of tests?
That debate goes on even at Ischemia Technologies."
The IMA test is now poised for wider distribution, and whether it will
be useful or not is something that clinicians will decide based on evidence
and experience, Dr. Christenson says.
At this point, Dr. Januzzi says, many ER physicians and cardiologists
are withholding judgment, and he thinks the large-scale trial is coming
at just the right time. "Physicians are waiting for these data," and when
the results are in, he predicts, there will be much wider recognition
of the value of the IMA assay.
Anne Paxton is a writer in Seattle. |
