A Novel Site-Targeted Ultrasonic Contrast Agent With Broad Biomedical Application
Background In this work, we report a novel targetable ultrasonic contrast agent with the potential to noninvasively define and localize myriad pathological tissues for diagnosis or therapy. The agent is a biotinylated, lipid-coated, perfluorocarbon emulsion that has low inherent echogenicity unless bound to a surface or itself.
Methods and Results In study 1, emulsions with and without biotin were suspended in buffered saline and imaged with a 7.5-MHz linear-array transducer. Neither emulsion manifested significant ultrasonic backscatter until avidin was added. Avidin-induced aggregation produced a marked enhancement in backscatter from the biotinylated but not from the control emulsion. In study 2, porcine fibrin clots in vitro were pretargeted with biotinylated antifibrin monoclonal antibodies and then exposed to avidin and then to biotinylated or control perfluorocarbon emulsions. The basal acoustic reflectivity of clots imaged with a 7.5-MHz linear-array transducer was uniformly low and was increased substantially by exposure to the targeted biotinylated emulsion. In study 3, partially occlusive arterial thrombi were created in dogs and then exposed to antifibrin antibodies and avidin in situ. Biotinylated or control emulsion was administered either in situ or systemically. At baseline, all thrombi were undetectable with a 7.5-MHz linear-array transducer. Thrombi exposed to antifibrin-targeted contrast exhibited increased echogenicity (P<.05); control thrombi remained acoustically undetectable.
Conclusions These data provide the first in vivo demonstration of a site-specific ultrasonic contrast agent and have potential for improved sensitivity and specificity for noninvasive diagnosis of thrombi and other pathological diseases.
Acoustic contrast agents, first introduced by Gramiak and Shah,1 have become an established tool for enhancing ultrasound and Doppler sensitivity. The preponderance of systemic contrast agents described to date function to enhance the appearance of the blood pool and to define its distribution and integrity.2 3 4 5 6 7 8 9 10 11 The concept of site-directed ultrasonic contrast agents has been discussed by many authors but, until recently, never demonstrated.
Unlike a blood pool agent, a site-directed ultrasonic contrast agent is designed to specifically and sensitively enhance the acoustic reflectivity of a pathological tissue that would otherwise be difficult to distinguish from surrounding normal tissue. Rapid expansion of the monoclonal antibody industry has fueled the recent emergence of site-targeted contrast agents by providing a plethora of specific and sensitive ligands directed against a wide spectrum of pathological molecular epitopes. As a result, site-targeted ultrasonic contrast agents are expected to broadly expand the diagnostic capability and utility of all clinical ultrasound modalities.
This report describes the discovery and development of a novel site-targeted acoustic contrast agent and demonstrates its utility for detection of vascular thrombi in vivo at clinically relevant ultrasonic frequencies.12 13 14 The acoustic contrast agent is a nongaseous, lipid-encapsulated, perfluorocarbon emulsion that is administered by a three-step approach (Fig 1⇓) that is based on the well-described avidin-biotin interaction.15 16 17 18 Given the broad spectrum of potential clinical applications, from imaging of tumors and thrombi to localizing adhesion molecules, this new agent may provide a versatile adjunct for better defining pathologies with ultrasound in numerous medical and surgical situations.
Preparation of Control and Biotinylated Perfluorocarbon Microemulsions
The biotinylated perfluorocarbon contrast agent was produced by incorporating biotinylated phosphatidylethanolamine into the outer lipid monolayer of a perfluorocarbon microemulsion. Briefly, the emulsion comprised perfluorodichlorooctane (40% vol/vol, PFDCO, Minnesota Manufacturing and Mining), safflower oil (2.0%, wt/vol), a surfactant comixture (2.0%, wt/vol), and glycerin (1.7%, wt/vol). The surfactant comixture included 64 mol% lecithin (Pharmacia Inc), 35 mol% cholesterol (Sigma Chemical Co), and 1 mol% N-(6-(biotinoyl)amino)hexanoyl)-dipalmitoyl-l-α-phosphatidylethanolamine (Pierce), which were dissolved in chloroform, evaporated under reduced pressure, dried in a 50°C vacuum oven overnight, and dispersed into water by sonication, resulting in a liposome suspension. The liposome suspension was transferred into a blender cup (Dynamics Corp of America) with perfluorodichlorooctane, safflower oil, and distilled, deionized water and emulsified for 30 to 60 seconds. The emulsified mixture was transferred to an S110 Microfluidics emulsifier and continuously processed at 10 000 PSI for 3 minutes. The completed emulsion was placed in stopper, crimp-sealed vials and blanketed with nitrogen until use. A control emulsion was prepared identically, except a nonbiotinylated phosphatidylethanolamine was substituted into the surfactant comixture. Biotinylated and control perfluorocarbon emulsion particle sizes were determined in triplicate at 37°C with a Brookhaven BI-90 laser-light-scatter, submicron-particle-size analyzer (Brookhaven Instruments Corp).
Validation of Biotinylated Phosphatidylethanolamine Incorporation
Biotinylated and control perfluorocarbon emulsions (20 μL) were added to 2.98 mL isotonic PBS, pH 7.4, and avidin in a polystyrene cuvette. Avidin (Pierce, Inc) was dissolved in PBS and was present in the cuvette to final concentrations of 0.0, 0.5, 1.0, 1.5, 2.0, and 2.5 μg/mL. All samples were prepared in duplicate, mixed by gentle inversion, and continuously agitated at low speed on a rotary table for 30 minutes at room temperature. Emulsion particle sizes were determined in triplicate at 37°C with a Brookhaven BI-90 laser-light-scatter, submicron-particle-size analyzer (Brookhaven Instruments Corp).
Measurement of Acoustic Scattering In Vitro From Biotinylated and Control Perfluorocarbon Emulsion Particles That Have Been Complexed With Avidin
Biotinylated and control (nonbiotinylated) perfluorocarbon emulsions were diluted in PBS (1.3 μL/mL PBS), placed within dialysis tubing (Spectra/Por 4, 25 mm, MWCO 12 000 to 14 000, Spectrum Medical Industries, Inc). Avidin (16.7 μg/mL) was added to each emulsion suspension and mixed by gentle inversion for 30 minutes. Emulsions within the dialysis tubings were ultrasonically imaged at 7.5 MHz within a PBS water bath at room temperature before and after the addition of avidin. Pixel gray-scale values were assessed in five temporally distinct, replicate freeze-frame images of each treatment group. Data were analyzed according to the model yij=ti+eij, where y is the pixel gray-scale level, t is the treatment group, and e is the error (SAS, Inc).
In Vitro Targeting of Porcine Plasma Clots
Whole porcine blood was obtained fresh and anticoagulated (9:1, vol/vol) with sterile sodium citrate. In a series of trials, plasma clots (nine) were produced by combining plasma and 100 mmol/L calcium chloride (3:1 vol/vol) with 5 U thrombin (Sigma Chemical Co) in a plastic tube through which a 5-0 suture was passed to provide a clotting surface. The plasma was allowed to coagulate slowly at room temperature. Half of the clots (five) were incubated individually with 150 μg biotinylated antifibrin monoclonal antibody (NIB 1H10)19 20 in 10 mL PBS with 1% crystalline BSA (Sigma Chemical Co) for 2 hours; the remaining clots (four) were maintained in PBS with 1% BSA. BSA was added during antibody incubations to minimize nonspecific protein binding to the polystyrene Petri dish walls. The antibody-treated clots were then incubated with excess avidin (50 μg/mL PBS) for 30 minutes, followed by biotinylated perfluorocarbon emulsion (30 μL/mL PBS) for 30 minutes. The control clots were treated similarly with control perfluorocarbon emulsion (30 μL/mL PBS). Targeted and control clots were retreated with avidin and biotinylated or control emulsion alone, respectively, to optimize surface saturation before ultrasonic interrogation. Ultrasonic (7.5-MHz) images were obtained before and after contrast exposure and were recorded onto super VHS videotape.
In Vivo Targeting of Canine Arterial Thrombi
Animals and Experimental Design
Thirteen arterial segments in 10 mongrel dogs were surgically exposed for electric induction of thrombus. In 6 femoral arteries, thrombi were created and exposed in situ to each component of site-targeted contrast system (ie, biotinylated antifibrin monoclonal antibody, avidin and biotinylated perfluorocarbon emulsion). In 3 femoral arteries, thrombi were exposed in situ to the control, nonbiotinylated perfluorocarbon emulsion. In 4 dogs, external carotid artery thrombi were induced and exposed to biotinylated antifibrin antibody and avidin in situ, and the targeted, biotinylated perfluorocarbon agent was administered systemically (1 intra-arterially, 3 intravenously). Inadvertent use of cathodal rather than anodal current prevented thrombus formation in 1 carotid artery study.
Femoral or external carotid arterial segments were exposed surgically for in situ induction of thrombi by electrical vascular injury and subsequent exposure to antifibrin monoclonal antibody–targeted perfluorocarbon contrast agent or the control perfluorocarbon emulsion. We21 and others22 have shown that electric vascular injury results in generation of a platelet-rich thrombus similar to that observed in the coronary arteries of patients with acute myocardial infarction. Animal protocols were approved by the Jewish Hospital and Washington University Animal Care committees.
Dogs weighing 20 to 30 kg were anesthetized with sodium pentobarbital (30 mg/kg IV), followed by 1% halothane in oxygen. The femoral or external carotid artery and branches were exposed. One proximal arterial branch was selected for cannulation. All other branches were ligated. The tip of a 23-gauge needle crimped on silver-plated copper wire was inserted obliquely into the artery and secured with 5-0 Prolene suture through connective tissue on either side. Anodal current (200 to 400 μA) was applied for 45 to 120 minutes to induce a partially occlusive thrombus. A Doppler flow probe was used to monitor the development of thrombus. Partial distal constriction of the artery was used to facilitate thrombus formation.
After thrombus formation, a 20-gauge catheter was inserted into the preserved proximal branch of the artery. Blood was flushed from the injured arterial segment with saline, and further blood flow was excluded by snares.
Administration of Control and Targeted Perfluorocarbon Contrast
For contrast-targeted thrombi, biotinylated antifibrin monoclonal antibody (150 μg NIB 5F3 or NIB 1H10 in 0.5 mL PBS, pH 7.2 to 7.4) was injected and incubated in the vessel for 1 hour. The distal snare was released, and unbound antibody was flushed away with 0.9% saline. The distal arterial occlusion was reestablished, and excess avidin (0.5 mg, Pierce) in PBS was applied for 30 minutes. Unbound avidin was flushed from the lumen, the distal arterial occlusion was reestablished in seven femoral arterial segments, and 0.2 mL biotinylated emulsion was administered for 30 minutes. In the femoral artery segments, targeted thrombi received a second exposure to avidin and biotinylated perfluorocarbon emulsion, respectively, to ensure thorough exposure of the thrombus. Control thrombi were created in three femoral artery segments, incubated with PBS in lieu of the antibody-avidin targeting system, and exposed twice to nonbiotinylated control perfluorocarbon emulsion.
Four dogs with external carotid thrombi were anticoagulated with heparin, and the targeted, biotinylated perfluorocarbon emulsion (5 mL) was injected as a bolus systemically rather than in situ. The targeted emulsion was administered into the common jugular vein in three dogs and into the common carotid in one.
Arteries were imaged after thrombus formation (baseline) and after each administration of antibody, avidin, and perfluorocarbon emulsions. The acoustically reflective needle electrode was used to localize regions of thrombosis for insonification. The presence of thrombus was confirmed in each dog by incision of the artery and/or intravascular ultrasound.
Analysis of Thrombus Before and After Contrast Administration
The subjective appearances of thrombus in five spatially and temporally independent frames obtained before and after each application of emulsion were determined by nine experienced, blinded observers. Detectability of echogenic material (thrombus) within the lumen of the vessel was scored as “no,” “unsure,” or “yes.” Eleven thrombi, 8 treated with the targeted perfluorocarbon contrast agent and 3 exposed to the control emulsion, were used. Images from the first dog were obtained with different gain and time-gain compensation settings and were excluded. Images from 1 control thrombus after two applications of emulsion were also excluded because of increased acoustic noise caused by the obvious presence of highly echogenic air bubbles from a saline drip flowing during the imaging. Frequency distributions of thrombus scores for images obtained from arteries exposed to control and targeted contrast agents were compared by χ2 analysis (SAS, Inc).
Ultrasonic Imaging, Recording, and Analysis
All ultrasonic imaging, in vivo and in vitro, used a 7.5-MHz linear, phased-array transducer with a Hewlett Packard 2500 Imaging System. Two-dimensional ultrasonic images were produced with fixed transmit, receive, and time-gain compensation levels and recorded onto super VHS videotape for playback and image analysis. Intravascular imaging was performed with a 30-MHz Sonicath cardiovascular intravascular ultrasound catheter (Boston Scientific Corp) in conjunction with a Hewlett Packard 100 Intravascular Imaging Console. Super VHS images were digitized with NuVista software (Truevision, Inc).
The appropriate and adequate incorporation of biotinylated phosphatidylethanolamine into the targeted emulsion was confirmed by determining the effect of increasing concentrations of avidin on apparent particle size for the control and biotinylated perfluorocarbon emulsion (Fig 2⇓). Control emulsion particles were 234±28 nm in diameter at baseline and remained unchanged in the presence of up to 2.0 μg avidin/mL (243±12 nm), indicating a lack of agglutination caused by the absence of the biotinylated lipid. Addition of avidin to the biotinylated emulsion resulted in a particle size increase from 263±9 to >2142±175 nm. Marked flocculation and sedimentation of the biotinylated perfluorocarbon emulsion were visible at avidin concentrations >2.0 μg/mL. These results document the small intrinsic particle size of the biotinylated perfluorocarbon emulsion, typically 1/10 that of other ultrasonic contrast agents, and demonstrate their appropriate coalescence in the presence of titrated concentrations of avidin.
The acoustic contrast effects of the biotinylated and control perfluorocarbon emulsions before and after exposure to avidin were delineated by suspending each in buffer within dialysis tubing. Although both suspensions were visually opaque, neither perfluorocarbon emulsion was evident above the ultrasonic image noise floor (Fig 3a and 3b)⇓. Addition of avidin clearly increased the acoustic reflectivity of the biotinylated (Fig 3d)⇓ but not of the control emulsion (Fig 3c)⇓. The average gray-scale levels (0=black, 255=white) of the control particles before (4.0±0.01) and after (4.0±0.01) avidin were similar (Fig 4⇓). Mean gray-scale level of biotinylated particles increased (P<.05) from 4.0±0.02 at baseline to 61.3±0.88 at 30 minutes after avidin. These findings demonstrate the low inherent acoustic reflectivity of the perfluorocarbon emulsion particles at clinically relevant ultrasonic frequencies. Although increased aggregate particle size of the biotinylated contrast may have contributed to the increased acoustic reflectivity observed, the enhancement of ultrasonic scattering was greater than expected and suggested that concentration of particles in a thin film could increase acoustic reflections of fluid-tissue interfaces.
The biotinylated perfluorocarbon emulsion was targeted in vitro to porcine plasma clots with the use of antifibrin monoclonal antibodies. Before exposure to emulsion, clots reflected little ultrasound at 7.5 MHz. Antifibrin monoclonal antibody targeting of the biotinylated perfluorocarbon emulsion clearly enhanced the acoustic reflections along clot perimeters, whereas the clots exposed to the control perfluorocarbon emulsion remained unchanged (Fig 5a and 5b)⇓. Analogous results were obtained with human plasma clots in vitro and with standard 3.5-MHz, two-dimensional, phased-array transducers (unpublished results).
The effectiveness of the targeted biotinylated contrast was demonstrated in vivo with arterial thrombi induced by electric, vascular injury and exposed to control or antifibrin monoclonal antibody–targeted perfluorocarbon emulsion. Before contrast administration, all acutely formed thrombi were poorly detected with the 7.5-MHz linear-array transducer (the Table)⇓. Thrombi treated with the control perfluorocarbon emulsion remained poorly detected by all observers after the first and second applications of nonbiotinylated contrast. The sensitivity for detecting thrombi treated with the targeted biotinylated contrast was increased (P<.05) after only one application. More than 96% of the still-frame thrombi images targeted with the perfluorocarbon acoustic contrast system were detectable after the second emulsion application. All carotid thrombi targeted by systemically administered biotinylated perfluorocarbon contrast were clearly detected after one exposure, and further acoustic enhancement by a second administration of avidin and emulsion was not required. A representative femoral artery thrombus before and after targeted contrast application is depicted in Fig 6⇓.
No thrombi were detected in arterial segments adjacent but distal to the site of vascular injury that were exposed to antibody, avidin, and biotinylated perfluorocarbon emulsion. Pixel gray-scale levels of contrast-enhanced thrombi (70.6±9.2) were greater (P<.05) than paired distal arterial lumen gray levels (0.6±0.3). Lack of distal binding further supports the well-documented specificity of the antifibrin monoclonal antibodies used and the resultant specificity of the targeted contrast system for thrombi. Moreover, no thrombi were detected at sites in the common carotid arteries of dogs in which external carotid thrombi were targeted systemically. In one dog, cathodal current (ineffective for inducing thrombi) was applied to injure the vessel, and no thrombus was detected ultrasonically with the targeted acoustic contrast or after resection and visual inspection of the vascular segment.
Two lots of antifibrin monoclonal antibodies were used in the present experiment that differed in biological activity owing to shipping circumstances. The first lot appeared to have lower biological activity per milligram in vitro and provided less acoustic enhancement of thrombi after a single application than the second lot. This diminished sensitivity was amplified by a second application of avidin and biotinylated perfluorocarbon emulsion. These results suggest that contrast reapplication may increase the sensitivity of detection when either the affinity or activity of the ligand is low or when the density of targeted molecular epitopes is sparse.
The production of specific site-targeted contrast agents remains a goal for every medical imaging modality. In this article, we describe the development and in vivo application of the first ligand-targeted contrast agent for ultrasound reported to date. The acoustic agent is a biotinylated, lipid-encapsulated perfluorocarbon emulsion that is applied in a triphasic manner to significantly enhance the acoustic reflectivity of specific tissues. In the present study, the targeted acoustic contrast was directed against thrombus; however, a wide variety of alternative clinical applications can readily be envisioned.
The specificity of the system is imparted by use of monoclonal antibodies that have been biotinylated and are pretargeted to the molecular epitopes of interest. In phase 2, avidin is administered that binds and cross-links antibodies to targeted tissues and forms immunocomplexes with unbound circulating antibodies that otherwise are rapidly cleared by the reticuloendothelial system. Cross-linking of the bound antibodies with avidin likely confers the advantages of increasing the avidity of the immunocomplex and providing a scaffold for the biotinylated perfluorocarbon emulsion to bind (phase 3). Sequential reapplication of avidin and biotinylated acoustic contrast serves to amplify the acoustic signal, if required, and may be used to increase the sensitivity of the contrast process if desired.
Other currently available ultrasonic contrast agents, such as microbubble technologies, are devoted to enhancement of the circulating blood pool.2 3 4 5 6 7 8 9 10 11 These agents appear to exhibit marked acoustic contrast effects from individual particles, and some appear to persist for long intervals in either the circulating volume or the tissues. Although the use of an acoustic contrast agent with high intrinsic acoustic scattering and a long circulating time would seem desirable for purposes of targeting, it may prove paradoxical that such particles could actually diminish the signal-to-noise ratio between the surface specular reflections of the bound, targeted particles and the circulating echoes of the unbound contrast agents. A potentially higher frequency of false-positive assessments may ensue.
Unlike highly echogenic, microbubble contrasts, the nongaseous perfluorocarbon-based emulsions have comparatively low acoustic reflectivity in circulation. In particular, the perfluorocarbon used in the present agent, when formulated at a particle size <400 nm, exhibits relatively poor circulating acoustic contrast properties. Previous research13 has revealed that targeted binding of perfluorocarbon emulsion particles to membrane surfaces, such as nitrocellulose, increases the bandwidth-delimited backscattered power (6 to 8 dB at 30 to 60 MHz) from an extremely thin film estimated to be <1 μm thick. The thickness of this layer cannot be resolved with radiofrequency analysis at 50 MHz (axial resolution <30 μm) from the specular reflection of the front wall of the membrane. Analogous thin films were deposited on thrombi and clots in the present experiment.
The physical mechanisms responsible for enhanced acoustic contrast are unclear at this time. Perfluorocarbon emulsions have a high mass density (≈1.9 times that of water) that probably affects the acoustic properties of individual particles when suspended in blood.23 However, the poor echogenicity of individual particles in emulsion and their marked enhancement of reflectivity when self–cross-linked or bound to a surface suggest additional mechanisms. Because the intensity of acoustic scattering from spherical particles generally increases according to the sixth power of their radius, some form of particle aggregation may provide a reasonable initial hypothesis for the augmented contrast effect. In addition, the degree of nonlinearity involved in the acoustic reflectivity of this site-targeted contrast agent is unresolved, and the potential role for harmonic imaging is presently unknown.
Potential Safety Issues
The acoustic particle is composed of lipids, generally recognized as safe, and perfluorocarbons. In general, perfluorocarbons are biologically inert and do not pose a toxicological risk from metabolic degradation.24 The chemical inertness, high biocompatibility, and excellent formulation stability of newer perfluorocarbon emulsions have made them desirable agents for parenteral uses, including artificial blood substitutes.25
Perfluorocarbon emulsions are cleared by phagocytosis through the reticuloendothelial system and ultimately eliminated transpulmonically during expiration.25 In some animals, not including humans or dogs, intravenous perfluorocarbon emulsions have been associated with increased pulmonary residual volume caused by pulmonary gas trapping.25 26 27 28 This is a reversible side effect that is most often associated with perfluorocarbons of high vapor pressure.24 The perfluorodichlorooctane used in the present study has relatively low vapor pressure. In those four anesthetized dogs that were administered the biotinylated perfluorocarbon contrast systemically, a brief 0 to 15 mm Hg decrease in peak systolic blood pressure was detected (unpublished results). However, these hemodynamic changes may be experimental artifacts because our test formulations were nonsterile and endotoxin content was not controlled.
Avidin-biotin interactions have been used to demonstrate the concept of targeted ultrasonic contrast. This approach has been previously described and used successfully in humans for immunoscintigraphy.15 16 17 18 Repeated use of avidin or streptavidin could lead to immunogenic reactions in some individuals. However, two potential solutions exist. Development of recombinant or organically synthesized avidin analogues with low antigenicity can be envisioned, given the extensive physical, chemical, and biochemical knowledge available about avidin-biotin interactions. Another approach could be to use an alternative targeting system that would permit amplification in a manner similar to that demonstrated with avidin-biotin.
Finally, the amount of monoclonal antibody required to produce adequate signal at the intended target site could potentially elicit toxic effects. The need for significant amounts of monoclonal antibody has previously been demonstrated by immunoscintigraphy studies, but the toxicity of this protein load is markedly diminished by “humanizing” the ligand or using F(ab) fragments.
The degree of acoustic reflectivity observed in the present in situ study represents the benchmark for future in vivo studies. Success will depend largely on the binding of adequate concentrations of antifibrin antibody to thrombus. Appropriate adjustment of the dosage of antibody given and the time allowed for its equilibration with the target should facilitate adequate ligand saturation of the target. The successful conjugation of avidin to biotinylated antibodies has been demonstrated in several human immunoscintigraphy protocols studies15 16 17 18 and would be expected to be repeatable in site-targeted contrast applications. Once cross-linked to antibodies, avidin would be expected to increase stability of the antibody-avidin complex to the target tissue and would be unlikely to dissociate from the biotinylated ligand.
In the present study, although no arterial segments revealed nonspecific binding of the antifibrin-targeted acoustic contrast in dogs in which thrombi were created and although no thrombus exposed was acoustically enhanced by nonspecific binding of the control emulsion, only one dog had no thrombus created and was exposed to the targeted acoustic contrast. In that dog, no evidence of thrombus was detected by ultrasonic interrogation or visual inspection of the excised arterial segments.
Because the contrast is a particle, access to nonvascular spaces may be limited. However, access by other routes, including intralymphatic, intraperitoneal, intrabiliary, transurethral, transbronchial, intrathoracic, or intrathecal, should broaden the potential clinical utility of this contrast agent.
Ligand-targeted perfluorocarbon ultrasonic contrast agents appear to have the potential to provide a powerful method for detecting any molecular epitope or receptor for which a biotinylated monoclonal antibody or ligand is available. The need for ionizing radiation could be eliminated in numerous clinical applications. Acoustic imaging of pathology enhanced with targeted perfluorocarbon contrast agents may now be possible with commercially available ultrasonic technology. Further investigation of the physical and biochemical properties of these agents should provide useful information that will enhance the prospects for the clinical applications of this new technology.
This work was supported by NIH grant HL-42950. Dr Wickline is an Established Investigator of the American Heart Association, Dallas, Tex. Dr Lanza is an American Heart Association, Missouri Affiliate fellow. We thank Drs Tracey A. Edgell and Sanj Raut at the National Institute for Biological Standards and Control, UK, for assistance in purification and preparation of the biotinylated antifibrin monoclonal antibodies.
- Received June 14, 1996.
- Revision received July 31, 1996.
- Accepted August 7, 1996.
- Copyright © 1996 by American Heart Association
Feinstein SB. New developments in ultrasonic contrast techniques: transpulmonary passage of contrast agent and diagnostic implications. Echocardiography.. 1989;6:27-33.
Mattrey RF, Wrigley R, Steinbach GC, Schutt EG, Evitts DP. Gas emulsions as ultrasound contrast agents: preliminary results in rabbits and dogs. Invest Radiol. 1994;29(suppl 2):S139-S141.
Goldberg BB, Liu J-B, Burns PN, Merton DA, Forsberg F. Galactose-based intravenous sonographic contrast agent: experimental studies. J Ultrasound Med.. 1993;12:463-470.
Unger E, Shen D, Fritz T, Kulik B, Lund P, Wu GL, Yellowhair D, Ramaswami R, Matsunaga T. Gas-filled lipid bilayers as ultrasound contrast agents. Invest Radiol. 1994;29(suppl 2):S134-S136.
Lanza GM, Wallace KD, Scott MJ, Sheehan CK, Cacheris WP, Christy DH, Sharkey AM, Miller JG, Wickline SA. Initial description and validation of a novel site targeted ultrasonic contrast agent. Circulation. 1995;92(suppl 1):I-260. Abstract.
Lanza GM, Wallace KD, Abendschein D, Scott MJ, Sheehan CK, Cacheris WP, Christy DH, Sharkey AM, Miller JG, Gaffney PJ, Wickline SA. Specific acoustic enhancement of vascular thrombi in vivo with a novel site-targeted ultrasonic contrast agent. Circulation. 1995;92(suppl I):I-260. Abstract.
Wallace KD, Lanza GM, Scott MJ, Holland MR, Christy DH, Sheehan CK, Cacheris WP, Gaffney PJ, Miller JG, Wickline SA. Initial description and validation of a novel site-targeted ultrasonic contrast agent. Circulation. 1995;92(suppl I):I-585. Abstract.
Paganelli G, Magnani P, Zito F, Villa E, Sudati F, Lopalco L, Rossetti C, Malcovati M, Chiolerio F, Seccamani E, Fazio F. Three-step monoclonal antibody tumor targeting in carcinoembryonic antigen-positive patients. Cancer Res.. 1991;51:5960-5966.
Dosio F, Magnani P, Paganelli G, Samuel A, Chiesa G, Fazio F. Three-step tumor pre-targeting in lung cancer immunoscintigraphy. J Nucl Biol Med.. 1993;37:228-232.
Modorati G, Brancato R, Paganelli G, Magnani P, Pavoni R, Fazio F. Immunoscintigraphy with three step monoclonal pretargeting technique in diagnosis of uveal melanoma: preliminary results. Br J Ophthalmol.. 1994;78:19-23.
Haskel EJ, Prager NA, Sobel BE, Abendschein DR. Relative efficacy of antithrombin compared with antiplatelet agents in accelerating coronary thrombolysis and preventing early reocclusion. Circulation.. 1991;83:1048-1056.
McGoron AJ, Pratt R, Zhang J, Shiferaw Y, Thomas S, Millard R. Perfluorocarbon distribution to liver, lung and spleen of emulsions of perfluorotributylamine (FTBA) in pigs and rats and perfluorooctylbromide (PFOB) in rats and dogs by 19F NMR spectroscopy. Artif Cells Blood Substit Immobil Biotechnol.. 1994;22:1243-1250.
Flaim SF, Hazard DR, Hogan J, Peters RM. Characterization and mechanism of side-effects of Imagent® BP (highly concentrated fluorocarbon emulsion) in swine. Invest Radiol.. 1991;26:S122-S124.
Clark LC, Hoffmann RE, Davis SL. Response of the rabbit lung as a criterion of safety for fluorocarbon breathing and blood substitutes. Biomat Art Cell Immobil Biotechnol.. 1992;20:1085-1099.