A Monoclonal Antibody That Recognizes the GPIIb/IIIa Antagonist DMP 728
Reversal of the Effects of DMP 728 on Platelet Aggregation and Bleeding Time in the Dog
Abstract Since hemorrhagic events represent a major safety concern associated with the use of new antithrombotic therapies such as glycoprotein (GP) IIb/IIIa receptor blockade, we evaluated the ability of a monoclonal antibody recognizing DMP 728 (cyclic [d-2-aminobutyryl-N2-methyl-l-argininyl-glycyl-l-aspartyl-3-aminomethyl-benzoic acid] methanesulfonic acid salt), a potent GPIIb/IIIa receptor antagonist, to reverse the pharmacological actions of DMP 728 in the dog. DC11 was chosen for in vivo evaluation based on its ability to inhibit the binding of [3H]DMP 728 to activated platelets and to attenuate the inhibition of ADP-induced aggregation on platelet-rich plasma ex vivo by DMP 728. After anesthesia mongrel dogs were given DMP 728 (20 μg/kg body wt IV) infused into the femoral vein, bleeding times were determined using a Simplate device from incisions on the backside of the tongue, and platelet aggregation was determined ex vivo. Nearly complete inhibition of platelet aggregation was observed for the dogs treated with DMP 728 (20 ug/kg IV) for up to 210 minutes, and bleeding times were prolonged >15 minutes for 2 hours and remained elevated for more than 4 hours. DC11 (0.2 or 1.0 mg/kg body wt IV) given to dogs 10 minutes after DMP 728 resulted in 50% attenuation of the effect of DMP 728 on aggregation at 3 hours. Approximately 34% inhibition of the DMP 728–mediated bleeding time was achieved at 1 hour with the 0.2 mg/kg dose, whereas approximately 50% inhibition of the bleeding time was observed for the 1 mg/kg dose at 1 hour. The bleeding time for the higher dose of DC11 returned to control levels at 2 hours, whereas the lower dose of DC11 returned to control levels at 3 hours. These results suggest that monoclonal antibody DC11 may have potential utility as an antidote to DMP 728 in neutralizing any unexpected bleeding complications associated with its use.
- Received June 8, 1995.
- Accepted October 11, 1995.
Clinical studies have provided conclusive prospective evidence associating platelet aggregability with the incidence of coronary artery disease.1 2 3 4 5 6 7 8 Platelets adhere at sites of vascular injury or damage, undergo activation, and express functional GPIIb/IIIa receptors, also referred to as αIIb/β3, for circulating adhesive ligands, principally fibrinogen.9 10 Formation of fibrinogen bridges via GPIIb/IIIa receptors on adjacent platelets results in the formation of platelet aggregates.
The GPIIb/IIIa receptor is an attractive therapeutic target for a number of different reasons. It is a platelet-specific protein and represents the most abundant platelet membrane protein, with approximately 50 000 copies expressed per cell.11 The binding of fibrinogen by functional GPIIb/IIIa receptors represents the final common step in platelet activation by all agonists.10 12 Recently, a human-murine chimeric Fab fragment of a monoclonal antibody directed against GPIIb/IIIa, designated 7E3, was evaluated by the EPIC investigators for use in patients.13 The results of the EPIC trial support the concept that GPIIb/IIIa receptor blockade is an effective strategy for the prevention of abrupt vessel closure after high-risk angioplasty and atherectomy. However, the EPIC trial also reinforced that, as with any new antithrombotic therapy, excessive bleeding is the predominant safety concern. The beneficial effect of 7E3 was achieved at the risk of a significant increase in bleeding complications and transfusions.13
DMP 728 is a selective GPIIb/IIIa receptor antagonist with potent antiplatelet and antithrombotic effects in animals14 15 16 that suggest its potential clinical utility in humans. In the present study we have explored the strategy of reversing the pharmacological actions of DMP 728 with the use of a specific antidote. Our approach involved the production of specific monoclonal antibodies recognizing DMP 728, characterization of their ability to neutralize the actions of DMP 728 in vitro, and evaluation of their effects on DMP 728–mediated platelet aggregation inhibition and bleeding time prologation in anesthetized dogs. Our results suggest that one monoclonal antibody, DC11, may represent an appropriate means to neutralize unexpected bleeding complications associated with the use of DMP 728.
DMP 728 and XL085 were synthesized at the DuPont Merck Pharmaceutical Company. RGDS was purchased from Sigma Chemical Co, human fibrinogen from Enzyme Research Laboratories, Inc., and human vitronectin from Calbiochem-Novabiochem Corporation. [3H]DMP 728 (24.3 Ci per mmol/L) was obtained from DuPont NEN.
Monoclonal antibodies were generated using standard hybridoma technology17 after one intrasplenic and three intraperitoneal immunizations of female BALB/c mice with XL085 conjugated to keyhole limpet hemocyanin. XL085 is a closely related analogue of DMP 728 with lysine residue that allows ready coupling to large carrier proteins. Following selection in Iscove’s modified Dulbecco’s medium (GIBCO) containing HAT media supplement (Boehringer Mannheim), hybridoma supernatants were screened for antibody by an ELISA with the use of polystyrene microtiter plates coated with 500 ng of XL085 conjugated to bovine serum albumin. Antibody-producing hybridomas were cloned by limiting dilution, and monoclonal antibodies were purified from culture supernatant by affinity chromatography over a Protein G Sepharose 4 Fast Flow column (Pharmacia). Class determination was performed with an isotyping ELISA (Bio-Rad). Cross-reactivity of monoclonal antibodies with DMP 728, linear RGDS, or the RGD-containing proteins fibrinogen and vitronectin was determined by preincubating the antibodies with test compound or protein before the XL085 ELISA, as previously described.15
Platelet-[3H]DMP 728 Binding Assays
The binding of [3H]DMP 728 to activated human platelets at room temperature was evaluated as follows. Human blood was collected in citrated Vacutainer tubes, centrifuged for 10 minutes at 100g, and the platelet-rich plasma (PRP) was carefully removed, making sure not to disturb the red blood cell pellet. PRP (205 μL; 4×108/mL, platelet concentration) was added to each well of a 96-well removawell plate (Dynatech), and the platelets were activated for 10 minutes with a mixture of agonists containing ADP, arachidonic acid, and epinephrine (100 μmol/L). Platelets were then incubated with [3H]DMP 728 (10 nmol/L) for 10 minutes, in the absence or presence of increasing concentrations of unlabeled DMP 728, by which time equilibrium binding had been established. To separate platelet-bound [3H]DMP 728 from free ligand, samples were centrifuged for 10 minutes at 1800g, and the supernatant was then removed. Platelet-bound [3H]DMP 728 was measured in a scintillation counter (Packard Minaxi Tricarb). Binding was analyzed by the method of Scatchard18 using nonlinear regression analysis methodology. To evaluate the effects of antibodies on [3H]DMP 728 binding, monoclonal antibodies were added to the activated platelets before [3H]DMP 728 was added. All binding results represent the mean of three separate experiments.
Platelet Aggregation Assays
The effect of DMP 728 on the aggregation of human platelets in response to 100 μmol/L ADP was determined as previously described.14 The ability of different antibodies to neutralize the inhibitory effects of DMP 728 on platelet aggregation was evaluated by pretreating the antibodies (100 nmol/L) with DMP 728 (100 nmol/L) for 10 minutes at 37°C before the aggregation assays.
Mongrel dogs of either sex weighing 8 to 12 kg were anesthetized with pentobarbital (30 mg/kg IV) (Nembutal, Abbott Laboratories). Additional anesthetic was administered as needed. Animals were placed on a respiratory pump (model 665, Harvard Apparatus), and the stroke volume of the respiratory pump was adjusted to the dog’s weight. The femoral artery was cannulated for blood collection, and the femoral vein was cannulated for compound administration. DMP 728 (20 ug/kg body wt IV) was administered to a group of dogs (n=6) followed 10 minutes later with the monoclonal antibody DC11 (0.2 or 1.0 mg/kg body wt IV). Blood samples were collected at various time points up to 5 hours for platelet aggregation determinations as previously described.14
Bleeding times were determined at those various time points using a Simplate device (Organon Teknika). Uniform incisions were made on the middle of the dorsal surface of the tongue, and the blood was blotted with filter paper at 30-second intervals until the bleeding stopped. If the bleeding did not stop within 15 minutes, a bleeding time of >15 minutes was assigned.
From each blood sample, 20 μL of PRP was added to 20 mL Isoton T diluent (Coulter Diagnostics), and platelet number was determined using a Coulter counter (Coulter Electronics, Inc).
From two different fusions using spleens from four XL085-immunized mice, 10 monoclonal antibodies were identified that recognized DMP 728 in the ELISA. Since DMP 728 is a cyclic RGD-containing molecule, antibodies were also tested for cross-eactivity against a linear RGD peptide (RGDS) and against the RGD-containing proteins, fibrinogen and vitronectin. None of the antibodies displayed appreciable cross-reactivity for RGDS, fibrinogen, or vitronectin, even at concentrations as high as 100 μg/mL. The antibodies were purified from culture supernatants for further studies.
[3H]DMP 728 Binding to Activated Platelets
Previous studies14 showed that DMP 728 inhibited platelet aggregation that was induced by ADP to a similar extent in dog and human platelets (IC50s of 15 and 46 nmol/L, respectively). In view of this similar sensitivity to DMP 728, we chose to characterize the binding of DMP 728 to human platelets. Scatchard analysis of [3H]DMP 728 binding to activated platelets is shown in Fig 1⇓. Nonlinear regression analysis of the curvilinear plot suggested the presence of two classes of binding sites: a high affinity site with a Kd of 0.07 nmol/L and a lower affinity site with a Kd of 9.9 nmol/L. This low affinity site could represent the binding of DMP 728 to unactivated platelets present in the preparation. Binding studies in which unactivated platelets were used support this interpretation, since a Kd of 30 nmol/L was determined. DMP 728 therefore displays a greater than 100-fold selectivity for activated versus unactivated platelets.
Once we had established the specific binding of DMP 728 to activated platelets, the effects of different antibodies recognizing DMP 728 on such binding were then determined. Six of the antibodies were effective in blocking DMP 728 binding in concentration-dependent fashion, with one IgG2a antibody in particular, designated DC11, identified as a potent competitor. The concentration-response curve for the DC11-mediated inhibition of DMP 728 binding to platelets is shown in Fig 2⇓. Under these assay conditions an IC50 of approximately 0.3 nmol/L was determined for DC11. A control IgG antibody, KAA8, which recognizes an irrelevant antigen angiotensin II,19 had no effect on [3H]DMP 728 binding, supporting the specificity of the DC11 effect (data not shown).
The antibodies were next evaluated for their ability to attenuate the inhibitory effect of DMP 728 on ex vivo ADP-induced platelet aggregation. Again, DC11 was a potent inhibitor of DMP 728 in this assay. Results shown in the Table⇓ indicate that 100 nmol/L DC11 was effective in blocking DMP 728 at either 20 or 100 nmol/L. As in the binding assay the KAA8 control antibody had no effect on DMP 728 activity (data not shown).
Effects of DC11 in Anesthetized Dogs Treated With DMP 728
The effects of DC11 on the pharmacological actions of DMP 728 were evaluated in anesthetized dogs. DMP 728, administered as a 20 μg/kg body wt IV bolus, elicited a complete inhibition of platelet aggregation induced by 100 μmol/L ADP; this effect was achieved within 5 minutes, and nearly complete inhibition was maintained for approximately 3 hours (Fig 3⇓). The effects of DC11, administered 10 minutes after the DMP 728 bolus, are also shown in Fig 3⇓. A decrease in the platelet-inhibitory effect of DMP 728 was noted after administration of DC11, at either a 0.2 or 1 mg/kg dose. This decrease reached statistical significance at the 140-minute time point. The higher dose of DC11 was slightly more effective in reversing the effects of DMP 728 on platelet aggregation, although by 3 hours the level of platelet inhibition was ≈50% for both doses. DC11 alone had no significant effect on platelet aggregation (data not shown). No significant effects on platelet counts were observed in any of the treatment groups (data not shown).
Fig 4⇓ shows the effects of DMP 728 in the absence and presence of DC11 on the Simplate-determined bleeding time in these anesthetized dogs. The 20 ug/kg IV bolus of DMP 728 increased bleeding times rapidly to greater than 15 minutes; this level of bleeding persisted for 2 hours; whereas by 5 hours bleeding time levels returned to near predosing values (≈3 minutes). When DC11 was administered 10 minutes after the DMP 728 bolus, ≈35% and 50% inhibitions of bleeding times at 1 hour were noted with the 0.2 and 1.0 mg/kg doses, respectively. These reductions reached statistical significance at the 60-minute time point. The bleeding time for the higher dose of DC11 returned to control levels at 2 hours, whereas the lower dose returned to control levels by 3 hours. A dissociation between the percent inhibition of platelet aggregation and bleeding time prolongation was noted. At 120 minutes there was a 50% reduction in bleeding time prolongation with the 0.2 mg/kg DC11 dose, although the inhibition of platelet aggregation was 85%. Complete inhibition of bleeding time prolongation was observed at 120 minutes for the high-dose DC11 group, in which ≈60% inhibition of platelet aggregation was observed. Previous studies have demonstrated that elevations of bleeding time with DMP 728 are observed only after levels of platelet aggregation reach 75% to 80%.14 15 16
The results of the EPIC trial provide strong support for the concept that GPIIb/IIIa receptor blockade is an effective strategy for the prevention of abrupt vessel closure after high-risk angioplasty and atherectomy.13 However, the beneficial effect of 7E3 was achieved at the risk of significant increases in bleeding complications and transfusions. Consequently, it would be desirable to have the appropriate means to neutralize any unexpected bleeding complications from 7E3 or any other pharmacological agent working through the mechanism of GPIIb/IIIa receptor blockade.
In this study, we have evaluated the strategy of the use of a specific antidote to neutralize the pharmacological actions of DMP 728, a potent and selective GP IIb/IIIa receptor antagonist with antiplatelet and antithrombotic effects that have been documented in various animal models.14 15 16 Our approach to the development of an antagonist to DMP 728 involved generating monoclonal antibodies that recognize DMP 728 and evaluating their ability to inhibit the functional activities of DMP 728 in vitro (namely, binding to platelets and inhibition of ADP-induced platelet aggregation). On the basis of its inhibitory effects on DMP 728 in these assays, antibody DC11 was tested to determine its ability to neutralize the pharmacological actions of DMP 728 in anesthetized dogs. DC11, administered 10 minutes after a dose of DMP 728 that completely abolished platelet aggregation and markedly prolonged Simplate-determined bleeding times, promoted a quicker return to predosing levels for both parameters. There was a suggestion of a dose-dependent effect for DC11; however, the rate of return to predosing levels for both parameters was not significantly different for the low-dose (0.2 mg/kg) and the high-dose (1.0 mg/kg) antibody groups. These data suggest that the affinity of the antibody for DMP 728 rather than its concentration is the principal determining factor in reversing the effects of DMP 728. In an earlier study we showed that DMP 728–mediated inhibition of platelet aggregation is related to plasma levels of the compound.14 This type of relationship would be predicted for a compound such as DMP 728, which displays a clear preference for activated rather than unactivated platelets. In addition, the volume of distribution at steady state for intravenous DMP 728 is low (0.8 L/kg), indicating there is very little penetration into the tissues. These observations support the notion that the DC11 mechanism of reversal entails complexing with free DMP 728 and promoting its clearance.
In view of the high affinity of DMP 728 for platelets (Kd=0.07 nmol/L) and its potency in inhibiting platelet aggregation, the ability of DC11 to reverse its pharmacological effects is striking. However, many questions remain concerning the potential clinical use of DC11, or any antibody antidote for a GPIIb/IIIa receptor antagonist. One issue is the rate at which DC11 reversed the effects of DMP 728, and whether this rate is fast enough to avert potentially severe hemorrhagic complications in clinical situations. It would be interesting to determine whether antibodies with higher affinity for DMP 728 could promote a more rapid reversal of the compound’s pharmacological effects than those observed with DC11.
A second important issue in this study is our use of Simplate-determined bleeding time as a measure of hemostatic function in the dog, and the relevance of this parameter for predicting bleeding complications in humans. Several investigators have pointed out that template bleeding time measurements in humans do not adequately predict hemorrhagic risks.20 21 22 Clinical evidence has shown that the bleeding time of the skin does not necessarily correlate with bleeding at other anatomic sites or at instrumented intravenous and intra-arterial access sites used for invasive procedures.23 Despite the controversy associated with the predictive risk value of bleeding time prolongations, the observation that DC11 inhibited the antiaggregatory effects of DMP 728 and its bleeding time effects indicate that the antibody is capable of reversing the pharmacological actions of the compound that could lead to unwanted side effects under certain conditions. Furthermore, our data suggest that there is a critical level of platelet inhibition that must be achieved with DMP 728 before prolonging the bleeding time.14 15 16 Full recovery of platelet aggregation is not required to correct the bleeding time.
In summary, these studies are to the best of our knowledge the first example that demonstrates the ability of a specific monoclonal antibody to neutralize in vivo the pharmacological actions of a GPIIb/IIIa receptor antagonist. Despite the fact that development of an antibody means a new drug that would have to be approved for use in patients, the results from this study suggest the clinical potential of antibody antidotes for new antithrombotic agents for which a major safety concern is abnormal bleeding.
Selected Abbreviations and Acronyms
|DMP 728||=||cyclic [d-2-aminobutyryl-N2-methyl-l-argininyl-glycyl-l-aspartyl-3-aminomethyl-benzoic acid] methanesulfonic acid salt|
|ELISA||=||enzyme-linked immunosorbent assay|
|EPIC||=||Evaluation of 7E3 for the Prevention of Ischemic Complications|
|XL085||=||cyclic [d-Lysyl-N2-methyl-l-argininyl-glycyl-l-aspartyl-3-aminomethyl-benzoic acid]|
Fuster V, Steele PM, Chesebro JH. Role of platelets and thrombosis in coronary atherosclerotic disease and sudden death. J Am Coll Cardiol. 1985;5:175B-184B.
Hennekens CH, Buring JE, Sandercock P, Collins R, Peto R. Aspirin and other antiplatelet agents in the secondary and primary prevention of cardiovascular disease. Circulation. 1989;80:749-756.
Trip MD, Cats VM, van Capelle FJL, Vreeken J. Platelet hyperreactivity and prognosis in survivors of myocardial infarction. N Engl J Med. 1990;22:1549-1554.
Thaulow E, Ericssen J, Sandvik L, Stormorken H, Cohn PF. Blood platelet count and function are related to total and cardiovascular death in apparently healthy men. Circulation. 1991;84:613-617.
D’Sousa SE, Ginsberg MH, Burke TA, Plow EF. The ligand binding site of the platelet integrin receptor GPIIb-IIIa is proximal to the second calcium binding domain of its α subunit. J Biol Chem. 1990;265:3440-3446.
Pytele R, Pierschbacher MS, Ginsberg MH, Plow EF, Ruoslahti E. Platelet membrane glycoprotein IIb/IIIa: member of a family of RGD specific adhesion receptors. Science. 1986;231:1559-1562.
Mousa SA, Bozarth JM, Forsythe MS, Jackson SM, Leamy A, Diemer MM, Kapil RP, Knabb RM, Mayo MC, Pierce SK, De Grado WF, Thoolen MJ, Reilly TM. Antiplatelet and antithrombotic efficacy of DMP 728, a novel platelet GPIIb/IIIa receptor antagonist. Circulation. 1994;89:3-12.
De Caterina R, Lanza M, Manca G, Strata GB, Maffei S, Salvatore L. Bleeding time and bleeding: an analysis of the relationship of the bleeding time test with parameters of surgical bleeding. Blood. 1994;84:3363-3370.