(Circulation. 1995;92:2373-2380.)
© 1995 American Heart Association, Inc.
Articles |
From the Department of Medicine, Mount Sinai School of Medicine, New York, NY.
Correspondence to Dr Barry S. Coller, Department of Medicine, Mount Sinai School of Medicine, 1 Gustave L. Levy Place, New York, NY 10029.
| Introduction |
|---|
|
|
|---|
I first review the evolution of our understanding of platelet physiology since the 1960s and why the platelet GPIIb/IIIa receptor appeared to be a logical target for antithrombotic therapy. I then describe our studies with 7E3 to block the receptor in experimental animal models of thrombosis. The results of these studies encouraged us to try to develop 7E3 as a human therapeutic agent. The development was a monumental effort conducted by hundreds of people working for nearly a decade and culminated in the approval by the Food and Drug Administration of c7E3 Fab (abciximab [ReoPro]) in December 1994 as conjunctive therapy for use in patients undergoing high-risk coronary artery angioplasty and atherectomy. A detailed description of the developmental process is beyond the scope of this essay, so it is only briefly described. My experience in moving back and forth from the academic medical center research bench to the world of drug development has allowed me to reflect on the processes involved, and at the end of this essay, I share some of my thoughts and concerns.
| The GPIIb/IIIa Receptor as a Therapeutic Target |
|---|
|
|
|---|
Glanzmann Thrombasthenia
Glanzmann thrombasthenia was first
described in
1918,1 but its modern description as a hereditary disorder
causing mucocutaneous hemorrhage, marked prolongation of the
bleeding time, and abnormal clot retraction dates from the
mid-1960s.2 3 4 Studies at that time
demonstrated that
patients' platelets failed to aggregate at all in response to all
of the agonists believed to operate in vivo, including ADP,
epinephrine, serotonin, collagen, and thrombin.
This profound defect in platelet aggregation contrasted with the
much more modest inhibition of platelet aggregation produced by
aspirin, whose discovery as an antiplatelet agent occurred at
about the same time.5 6 Patients with Glanzmann
thrombasthenia have serious mucocutaneous bleeding, but despite having
essentially infinite bleeding times throughout their lives, spontaneous
gastrointestinal and genitourinary hemorrhage occurs only
sporadically, and spontaneous central nervous system hemorrhage
is exceedingly rare.7 8
One important clue to both the pathogenesis of Glanzmann disease and the mechanism by which normal platelets aggregate was the observation made independently by Zucker et al4 and Jackson et al9 that Glanzmann patients' platelets are severely deficient in fibrinogen. In the mid-1970s, Nurden and Caen10 and Phillips et al11 independently identified deficiencies of two different platelet membrane glycoproteins in several kindreds with Glanzmann thrombasthenia. These glycoproteins were designated glycoprotein (GP) IIb and GPIIIa based on their relative electrophoretic mobilities in sodium dodecyl sulfatepolyacrylamide gel electrophoresis. Soon thereafter, it was established that these two glycoproteins exist as a calcium-dependent complex (GPIIb/IIIa) on the platelet surface.12
Platelet-Fibrinogen Interactions
It took until the late 1970s
to establish that the binding of
fibrinogen to the platelet surface is necessary for normal in vitro
platelet aggregation induced by all of the agonists believed to
operate in vivo.13 14 15 Although the
GPIIb/IIIa receptor
immediately became a prime candidate as the platelet "fibrinogen
receptor," controversy persisted as to whether the abnormality in
Glanzmann thrombasthenia was due to a deficiency in the
platelet activation machinery or in the receptor
itself.16 Studies that I conducted on platelet
interactions with immobilized fibrinogen in 1980 supported
the view that the defect was in the receptor itself,17 and
data from others showing direct in vitro interactions between
GPIIb/IIIa and fibrinogen lent support to this view.18 The
technology for preparing monoclonal antibodies using murine hybridomas
was becoming generally available at this time, and we and others began
to try to develop such antibodies against platelet
glycoproteins.19 20 21 22
Because we wanted to
conduct functional studies, our strategy was to restrict our search to
antibodies that would interfere with the interaction between
platelets and immobilized fibrinogen. We were able to
produce such antibodies, and studies with one of them (10E5) showed
that it could abolish platelet aggregation of normal platelets,
block platelet-fibrinogen interactions, and inhibit clot
retraction; that is, it could essentially induce a functional
thrombasthenic phenotype.20 This antibody, and
others with similar properties prepared by other
investigators,19 22 immunoprecipated both GPIIb and
GPIIIa, providing crucial confirmatory evidence that these
glycoproteins exist as a complex and are involved in
fibrinogen binding. Approximately 40 000 antibody molecules bind to
the surface of platelets, indicating that there are probably
40 000 to 80 000 GPIIb/IIIa receptors per platelet, depending on
whether the antibodies bind bivalently or monovalently. GPIIb/IIIa thus
is probably the most dense adhesion/aggregation receptor present on
any cell.
Integrin Receptors
During the 1980s, it became apparent that
investigators studying a
wide variety of cell-protein and cell-cell interactions
involved in developmental biology, cellular differentiation, metastasis
formation, leukocyte function, and trafficking of cells in the immune
system were studying different members of a large family of receptors,
ultimately termed integrins.23 24 All of these
receptors
are composed of two chains, an
subunit and a ß subunit, which are
held together by noncovalent bonds. Both
and ß subunits are
transmembrane proteins. The
subunits characteristically have three
or four divalent cation binding domains, whereas ß subunits contain
many disulfide bonds. The cloning of the cDNAs for GPIIb and
GPIIIa25 26 led to the identification of GPIIb as a
typical
integrin subunit (
IIb) and GPIIIa as a
typical ß subunit (ß3). The GPIIb/IIIa receptor is
platelet specific and thus uniquely adapted for platelet
function. GPIIIa (ß3) however, can form a complex with
another
subunit,
v, to form the
vß3 vitronectin
receptor,27 which is present on
endothelial cells, osteoclasts, and other cells; trace
amounts of
vß3 are also present on
platelets (
50 to 100 receptors per
platelet).28
The RGD Cell Binding Mechanism
Investigators in the early
1980s studying the
5ß1 "fibronectin" integrin receptor
made a striking discovery concerning the mechanism by which fibronectin
binds to this receptor. Using proteolytic digestion fragments of
fibronectin and synthetic peptides comprised of amino acids sequences
contained within the fibronectin sequence, they were able to identify a
threeamino acid sequence, arginine-glycine-aspartic acid
(single letter code, RGD), that was crucial for the
binding.23 At approximately the same time, we and others
showed that von Willebrand factor, fibronectin,
vitronectin, and thrombospondin could all bind to
platelet GPIIb/IIIa under appropriate conditions of platelet
activation and that the binding could be blocked by small synthetic
peptides containing the RGD
sequence.29 30 31 32 33 34
Also at about
the same, the cDNAs of the adhesive glycoprotein ligands
were cloned and their amino acid sequences were deduced, yielding the
remarkable discovery that RGD sequences were present in all of
these GPIIb/IIIa ligands. Thus, the promiscuity of GPIIb/IIIa in
binding multiple ligands was explainable on the basis of a common
binding mechanism involving the RGD sequence. Ironically, although
fibrinogen contains two pairs of RGD sequences in its A
chains, it
appears that a sequence at the carboxyl terminus of the
chain is
responsible for its initial binding to GPIIb/IIIa.35 This
sequence has similarities to RGD, however, and the binding of peptides
based on this sequence to platelets is also inhibited by
RGD-containing peptides.36 Thus, it is likely that the
chain sequence binds to the RGD site or a related site on
GPIIb/IIIa.
Although many different proteins can bind to GPIIb/IIIa in vitro under appropriate conditions of platelet activation, during ex vivo platelet aggregation in normal plasma, fibrinogen appears to be the dominant protein binding to the receptor.31 Studies with ex vivo flow chambers designed to simulate better in vivo conditions, however, indicate that von Willebrand factor plays a more important role than fibrinogen.37 No direct data on the relative contributions of these glycoproteins to platelet aggregation in vivo are available, and the roles, if any, of the other adhesive glycoproteins in platelet adhesion and aggregation remain to be established.
A Model of Platelet Adhesion and Aggregation
Collectively,
the data presented provide the basis for our
current views of platelet physiology (Fig 1
). Thus,
damage to a normal blood vessel or an atherosclerotic plaque results in
exposure of adhesive glycoproteins such as von
Willebrand factor and collagen. Platelets have receptors
for these glycoproteins (Table
) that result
in platelet adhesion. These receptors are present on the
surface of platelets in their active states so that they can effect
adhesion immediately after vascular injury. Thus, platelet adhesion
is controlled by the normal endothelial lining
"hiding" the adhesive glycoproteins from the
circulating platelets.
|
|
After platelet adhesion, the platelet undergoes an activation process that produces a conformational change in the GPIIb/IIIa receptor so that it has a high binding affinity for fibrinogen, von Willebrand factor, and perhaps other glycoproteins. Binding of fibrinogen and/or von Willebrand factor to GPIIb/IIIa receptors is followed by other events that probably involve receptor clustering. Because both fibrinogen and von Willebrand factor are multivalent molecules, they can bind to GPIIb/IIIa receptors on two different platelets simultaneously, resulting in platelet crosslinking and platelet aggregation.
From a teleological
standpoint, the aggregation mechanism probably
reflects the need for a rapid and effective response to
hemorrhage and, thus, the high density of GPIIb/IIIa on the
surface of platelets and the high concentration of circulating
fibrinogen. If the resting GPIIb/IIIa receptor had a high affinity for
fibrinogen and/or von Willebrand factor, however, platelet
thrombus formation would occur continuously and produce inappropriate
thrombosis. Thus, control over platelet aggregation is exercised by
requiring the presence of one or more platelet-activating
agents, most of which are synthesized and/or released only at sites of
vascular injury (Fig 2
). There are positive feedback
loops in the system; activated platelets synthesize and/or
release the platelet activators thromboxane
A2, ADP, and serotonin, and they also
facilitate thrombin formation. It is notable that
thrombolytic agents can also activate
platelets, either directly or indirectly via thrombin
generation,38 39 suggesting that the net clinical
effect
of these agents involves competition between their fibrinolytic effects
and their ability to enhance platelet deposition. Finally, high
shear forces have been shown to activate platelets ex
vivo,40 and it is possible, but not certain, that such
forces can be achieved in vivo.
|
It is likely that combinations of agonists are present in vivo at the sites of vascular injury and that the mixture of agonists differs in different thrombi, or even differs over time in a single thrombus. It is notable, therefore, that combinations of agonists can show additive or synergistic effects on platelet activation.41 Thrombin appears to play a central role in some models of platelet thrombus formation but not others.42 43 44 Its precise role in thrombus formation in humans with vascular injury to an atherosclerotic plaque remains to be defined.
The membrane signals induced by the agonists are transduced by several mechanisms, including arachidonic acid metabolism and protein kinase C activation, but our understanding of these mechanisms remains incomplete, and it is certain that other mechanisms are also involved.41 Because aspirin blocks only arachidonic acid metabolism, it is only a partial inhibitor of platelet aggregation45 and offers only incomplete protection from platelet-mediated thrombosis.46 The final common pathway for platelet aggregation, regardless of agonist, is the conformational change in the GPIIb/IIIa receptor that results in it developing high affinity for its adhesive glycoprotein ligands.
Platelets in Human and Experimental Vascular
Disease
An abundance of evidence derived from examination of
pathological
specimens from patients with ischemic vascular events
implicated platelet aggregation as a major initiating
factor,47 48 and biochemical evidence of platelet
activation during ischemic vascular events provided independent
support for this view.49 Animal models of ischemic
vascular disease also incriminated platelets in initiating,
supporting, and exacerbating the thrombotic
process.42 43 44 50 51 52
They also identified that virtually no
vaso-occlusion or ischemic damage results from just a
monolayer of platelets, and thus platelet adhesion in the
absence of platelet aggregation is not likely to produce
ischemic damage.
GPIIb/IIIa as a Therapeutic Target
Analysis of all of these
data suggested that blockade of
GPIIb/IIIa receptors might be a desirable therapeutic strategy because
the monoclonal antibodies to GPIIb/IIIa are more potent
inhibitors of platelet function than aspirin,
GPIIb/IIIa is platelet specific, inhibition of GPIIb/IIIa still
leaves platelet adhesion largely intact and platelet adhesion
may contribute to hemostasis without leading to ischemic
damage, and the hemorrhagic diathesis produced by the inherited
deficiency of GPIIb/IIIa receptors found in Glanzmann
thrombasthenia only rarely produces spontaneous central nervous system
hemorrhage, the most feared complication of anticoagulant and
antiplatelet therapy.
| Development of 7E3 as a Therapeutic Agent |
|---|
|
|
|---|
Preliminary Canine Studies
As a prelude to conducting in vivo
studies, we had to produce
sizable amounts of purified 7E3 free of endotoxin. We also decided to
cleave off its Fc region [producing an F(ab')2
fragment]
so as to eliminate the possibility that platelets coated with the
intact antibody would be rapidly cleared in the spleen and elsewhere by
macrophages bearing Fc receptors. Armed with only a limited
supply of 7E3-F(ab')2, we treated three dogs at the
State University of New York at Stony Brook with increasing doses,
ultimately achieving virtually complete inhibition of platelet
aggregation, a feat that had not been reported previously for any
antiplatelet agent administered in vivo.54 The
animals did not develop spontaneous hemorrhage, nor did they
demonstrate any other gross evidence of toxicity.
GPIIb/IIIa Receptor Blockade Assay
To correlate the
functional affects of 7E3 treatment with the
percentage of GPIIb/IIIa receptors blocked, we developed an ex vivo
assay using the binding of radiolabeled 7E3 to platelets obtained
before and after 7E3 treatment to directly assess the percentage of
receptors blocked. Using this assay, we were able to show that at
antibody doses that decrease the number of available receptors to less
than 50% of normal, platelet aggregation shows significant
inhibition. At
80% GPIIb/IIIa receptor blockade, platelet
aggregation is nearly completely eliminated, but the bleeding time is
only mildly affected. It is only with >90% receptor blockade that the
bleeding time becomes >15 to 30 minutes.55 56
Quantitative analysis of 7E3-F(ab')2 binding was
also helpful in deciding whether to proceed with development of 7E3 as
a therapeutic agent as opposed to trying to obtain another antibody of
higher affinity. When doses of 7E3-F(ab')2 that produced
near-saturation of GPIIb/IIIa receptors were administered to
primates,
60% of the injected antibody actually became bound to
platelets. Compared with other drugs, this percentage binding to
the target receptor is remarkably high, indicating that the effective
dose is less than twice the theoretical minimum required to occupy the
available receptors. Thus, it was unlikely that we could produce a
"better" antibody, regardless of differences in affinity measured
in vitro.
Animal Models of Thrombosis
The greater potency of 7E3 in
both vitro and in vivo in inhibiting
platelet function compared with aspirin made us eager to test it in
animal models of thrombosis. Deciding on which animal models to test
was difficult since a large number of models existed and our supply of
antibody was limited. We considered two criteria to be most important:
(1) the model had to convincingly simulate a human vascular disease,
and (2) aspirin had to have failed to show complete protection. The
latter criterion guaranteed that we would be able to assess whether
7E3-F(ab')2 treatment was more effective than aspirin.
Dr John Folts had developed a model of cyclical flow reductions in the coronary artery of the dog and carotid artery of the monkey designed to simulate the vascular abnormalities found in patients with unstable angina and transient ischemic attacks, respectively.42 The model involves creating a partial stenosis of the vessel with an external cylindrical device and damaging the vessel with a hemostat. Platelet thrombi then form, resulting in a decrease in blood flow. The thrombi subsequently embolize several minutes later, either spontaneously or as a result of shaking the cylinder, and blood flow is restored. A new cycle is then initiated, and the cycles keep repeating, causing cyclical flow reductions. Aspirin can diminish or abolish the cyclical flow reductions, but the cyclical flow reducations can be restored by infusing a low dose of epinephrine or increasing the stenosis.46 Willerson and colleagues43 57 made substantial contributions to this model, defining the roles of serotonin, thromboxane A2, and thrombin in platelet thrombus formation. Most important, they demonstrated the clinical relevance of the model by showing that cyclical flow reductions occur in humans with atherosclerotic vascular disease, in particular, after coronary angioplasty.58 59
One of the most exciting days of my research career was my first visit to Dr Folts' laboratory when I was able to directly watch the effect of the first dose of 7E3-F(ab')2 in this open-chest model. The rapid improvements in the color and contractility of the heart after 7E3-F(ab')2 injection were dramatic, offering hope that similar phenomena might occur in humans with acute ischemic events.
7E3-F(ab')2 was the most potent agent that Dr Folts had tested in his model, inhibiting cyclical flow reductions despite severe prothrombotic provocations such as infusing epinephrine, increasing the vascular stenosis, increasing the vascular damage, and even passing current through the stenosing cylinder.56 60 Electron microscopic examination of blood vessels from treated animals showed only a single monolayer of platelets, despite the presence of extensive vascular damage. Dose-titration studies demonstrated that to achieve an antithrombotic effect in this model, it was not necessary to obtain very high grade GPIIb/IIIa receptor blockade, and thus prolongation of the bleeding time was only modest.55 Our studies with Dr Folts anticipated the later human studies by Anderson et al,58 59 who demonstrated that 7E3 eliminates the cyclical flow reductions that occurr in some patients after angioplasty.
Dr Chip Gold and his colleagues at the Massachusetts General Hospital were interested in the mechanism of coronary artery reocclusion after successful thrombolysis with recombinant tissue-type plasminogen activator (rTPA) and had developed a dog model of this phenomenon by placing a severe stenosis in series with a fresh thrombus in the left anterior descending coronary artery.50 Thrombolysis with rTPA could be achieved in nearly all animals, but reocclusion usually occurred within minutes, with many animals going on to develop cyclical flow reductions like those seen in the Folts model. Aspirin provided at best only a partial benefit in a fraction of the animals. My first visit to Dr Gold's laboratory was another landmark event for me since the coronary artery of a dog pretreated with 7E3-F(ab')2 did not reocclude after treatment with rTPA, much to the amazement of the skeptical onlookers.
Subsequent studies by us and other researchers further demonstrated that pretreatment with 7E3-F(ab')2 not only led to full protection against vascular reocclusion but actually shortened the time to reperfusion and allowed for dramatic reductions (up to 75%) in the dose of rTPA needed to achieve reperfusion; in fact, reperfusion occurred in some animals after just receiving 7E3-F(ab')2, even without getting rTPA.50 61 Moreover, platelet-rich thrombi that could not be lysed by rTPA alone were readily lysed when animals were treated with 7E3 before the rTPA was given.52 To explain these observations, we hypothesized that because thrombolytic agents can activate platelets,38 perhaps potent inhibition of platelet aggregation by 7E3 allows the thrombolytic effects of both exogenous rTPA and endogenous TPA to act unopposed by increased platelet deposition.62
These series of animal experiments and the independent confirmation of the antithrombotic effects of 7E3 treatment in other models by Lucchesi, Mickelson, Bates, Rote, and their colleagues,51 63 64 65 as well as other laboratories,66 67 68 formed the theoretical basis for our decision to move forward with the development of 7E3 as a therapeutic agent. Our data suggested that it may be of benefit in treating patients with unstable angina, transient ischemic attacks, and myocardial infarction, as well as in the prevention of complications of angioplasty. What remained unknown was whether the animal models of thrombosis really simulated the human disease and whether the hemorrhagic risks in humans would be acceptable.
Licensing 7E3 to Centocor and Development of 7E3 as a
Drug
The next steps in drug development could not be performed in my
laboratory because it would require resources far in excess of those in
my grant from the National Heart, Lung, and Blood Institute to study
basic platelet physiology. As a result, in 1986 the Research
Foundation of the State University of New York licensed 7E3 to
Centocor, Inc, a new biotechnology company specializing in the
diagnostic and therapeutic application of monoclonal
antibodies.
The subsequent development of 7E3 required extensive
collaboration
among myself, a large number of outstanding scientists at Centocor, and
many leading academic cardiologists, resulting in the preparation for,
the design, and the conduct of the clinical trials (see
"Acknowledgments"). Space, unfortunately, does not permit
discussing the many decisions and hurdles that remained for us,
including the choice to use the 7E3 Fab fragment and ultimately to
develop a mouse/human chimeric 7E3 Fab (c7E3 Fab)69 ; the
design and execution of the toxicology studies; the assessment of 7E3
crossreactivity with
vß3 and perhaps
activated MAC-1 receptors56 ; the development of
sensitive and specific assays to assess immune responses to c7E3 Fab;
the design, execution, and analysis of the phase I, II, and III
studies; and the preparation, submission, and defense of the
Product Licensing Application to the Food and Drug Administration
and comparable documents to European and Scandinavian agencies.
Based on the results of the 2099 patient EPIC trial,70 71 in which conjunctive treatment with a bolus plus infusion of c7E3 Fab significantly reduced the risk of developing an ischemic complication (death, myocardial infarction, or need for urgent intervention) after coronary artery angioplasty or atherectomy in patients at high risk of such complications, the Food and Drug Administration approved the conjuctive use of c7E3 Fab (generic name, abciximab; trade name, ReoPro) in high-risk angioplasty and atherectomy on December 22, 1994. We are continuing to explore the basic mechanisms by which abciximab works, including potential effects on thrombin formation72 and restenosis; we are also assessing changes in the heparin dosing to decrease the hemorrhagic risk associated with abciximab treatment,73 potential new indications for abciximab, and the safety and efficacy of repeat administration of abciximab.
| Reflections |
|---|
|
|
|---|
1. The role of the National Institutes of Health. All of the basic work in my laboratory and much of the work conducted in other laboratories that led to the choice of the GPIIb/IIIa receptor as a therapeutic target was funded by the National Heart, Lung, and Blood Institute. This funding was based on the importance of the scientific questions, including the study of a very rare platelet disorder, and not the immediate prospects of new therapy for the most common cause of death in the United States. It is crucial that the public appreciate the vital role that National Institutes of Health funding of basic research has had on the development of new therapies.
2. The role of the State University of New York at Stony Brook. The investment by the State of New York in developing Stony Brook as a medical school and a research university of excellence made it possible for me to perform my research as a junior faculty member in an enriched scientific environment. By funding a hybridoma core facility in the Department of Microbiology under the direction of Dr Arnold Levine, junior faculty like myself were able to gain early access to this technology at a reasonable cost.
3. The role of the biotechnology and pharmaceutical
industries. If there were no mechanism to license 7E3 to a
biotechnology company, c7E3 Fab would never have been developed as a
therapeutic agent. Thus, technology transfer policies and experts play
a crucial role in the process of translating basic research into
diagnostic and therapeutic agents. Similarly, many fewer
drugs will be developed if the investors in biotechnology and
pharmaceutical companies do not believe that at the end of the 10 to 12
years it takes to develop a new drug that they will profit from their
exceedingly expensive (estimated at $359 million for research and
development per drug) and risky (only
1 of 10 novel agents that
reach phase I studies become approved drugs)
investment.74
4. Bottlenecks in translation of basic knowledge into new therapies. The daunting economic statistics cited above result in biotechnology and pharmaceutical companies choosing only a small number of agents to develop. Thus, it may be very difficult for investigators to develop licensing agreements for agents that may have limited markets or seem very risky. Without such agreements it is difficult for individual investigators to get over the hurdles of preparing purified materials suitable for in vivo use and developing suitable animal models to assess potential human safety and efficacy. Thus, support of resources designed to help investigators overcome these hurdles would likely lead to an extraordinary amount of important pathophysiological information and perhaps new drugs. A commitment to supporting individuals trying to develop and assess clinically relevant animal models through the techniques of experimental pathology is also likely to speed the process of translating basic knowledge into clinical benefit. The explosion in transgenic mouse models of disease reinforces this point.
5. Scale-up from the academic environment to the
biotechnology and pharmaceutical level is breathtaking and exceedingly
difficult. The first functional screening assay performed with 7E3
required approximately 1 millionth of a gram (ie,
1 µg) of protein
(100 µL of hybridoma culture supernatant containing
10 µg/mL of
antibody). Abciximab is now produced in lots of more than 1000 g of
protein, which is more than 1 billion times the amount that was
initially used. Academic scientists rarely have the chance to learn
just how difficult it is to develop scale-up technology and meet
the stringent criteria established by the Food and Drug Administration
for good manufacturing practices of a pharmaceutical.
6. Need for physician and public education. Despite extensive courses in pharmacology during medical school, most medical students, trainees, and practicing physicians learn very little about the process of drug development. It is not surprising then that the public also knows so little. If the public debate about the costs of medications and the appropriate incentives for developing new drugs is to be reasoned, public information is crucial. Finally, it is vital that the public understand the mission of the Food and Drug Administration, the rigor required to meet the standards for safety and efficacy required for new drug approval, and the Food and Drug Administrationimposed restraints on marketing practices.75 76 Armed with this knowledge, the public will be better able to compare the quality and quantity of the safety and efficacy data available on Food and Drug Administrationapproved drugs with those available on "natural remedies."
| Acknowledgments |
|---|
| References |
|---|
|
|
|---|
This article has been cited by other articles:
![]() |
P. Monagle, E. Chalmers, A. Chan, G. deVeber, F. Kirkham, P. Massicotte, and A. D. Michelson Antithrombotic Therapy in Neonates and Children: American College of Chest Physicians Evidence-Based Clinical Practice Guidelines (8th Edition) Chest, June 1, 2008; 133(6_suppl): 887S - 968S. [Abstract] [Full Text] [PDF] |
||||
![]() |
G. Davi and C. Patrono Platelet Activation and Atherothrombosis N. Engl. J. Med., December 13, 2007; 357(24): 2482 - 2494. [Full Text] [PDF] |
||||
![]() |
B. Salanova, M. Choi, S. Rolle, M. Wellner, F. C. Luft, and R. Kettritz beta2-Integrins and Acquired Glycoprotein IIb/IIIa (GPIIb/IIIa) Receptors Cooperate in NF-{kappa}B Activation of Human Neutrophils J. Biol. Chem., September 21, 2007; 282(38): 27960 - 27969. [Abstract] [Full Text] [PDF] |
||||
![]() |
O. Ben-Yehuda Upstream/Downstream: Glycoprotein IIb/IIIa in Non-ST-Segment Elevation Myocardial Infarction J. Am. Coll. Cardiol., February 7, 2006; 47(3): 538 - 540. [Full Text] [PDF] |
||||
![]() |
P. Monagle, A. Chan, P. Massicotte, E. Chalmers, and A. D. Michelson Antithrombotic Therapy in Children*: The Seventh ACCP Conference on Antithrombotic and Thrombolytic Therapy Chest, September 1, 2004; 126(3_suppl): 645S - 687S. [Abstract] [Full Text] [PDF] |
||||
![]() |
S. Larrucea, C. Gonzalez-Manchon, N. Butta, E. G. Arias-Salgado, L. Shen, M. S. Ayuso, and R. Parrilla Agonist-induced aggregation of Chinese hamster ovary cells coexpressing the human receptors for fibrinogen (integrin alpha IIbbeta 3) and the platelet-activating factor: dissociation between adhesion and aggregation Blood, April 15, 2002; 99(8): 2819 - 2827. [Abstract] [Full Text] [PDF] |
||||
![]() |
G. J. Despotis, M. S. Avidan, and C. W. Hogue Jr Mechanisms and attenuation of hemostatic activation during extracorporeal circulation Ann. Thorac. Surg., November 1, 2001; 72(5): S1821 - 1831. [Abstract] [Full Text] [PDF] |
||||
![]() |
E. Houdart, F. Woimant, R. Chapot, C. Mounayer, C. Soria, and J.-J. Merland Thrombolysis of extracranial and intracranial arteries after IV abciximab Neurology, June 12, 2001; 56(11): 1582 - 1584. [Abstract] [Full Text] [PDF] |
||||
![]() |
J. A. Cairns, P. Theroux, H. D. Lewis Jr., M. Ezekowitz, and T. W. Meade Antithrombotic Agents in Coronary Artery Disease Chest, January 1, 2001; 119(1_suppl): 228S - 252S. [Full Text] [PDF] |
||||
![]() |
P. Monagle, A. D. Michelson, E. Bovill, and M. Andrew Antithrombotic Therapy in Children Chest, January 1, 2001; 119(1_suppl): 344S - 370S. [Full Text] [PDF] |
||||
![]() |
P. J. Goldschmidt-Clermont, G. E. Cooke, G. M. Eaton, and P. F. Binkley PlA2, a variant of GPIIIa implicated in coronary thromboembolic complications J. Am. Coll. Cardiol., July 1, 2000; 36(1): 90 - 93. [Full Text] [PDF] |
||||