CA2178487C - Local delivery of fibrinolysis enhancing agents - Google Patents

Abstract

A method of preventing adhesions by topical administration of fibrinolysis e nhancing agents is described. The method uses a topicall y applied polymeric matrix for delivery of a fibrinolytic agent, preferably ur okinase, tPA, hirudin, or most preferably, ancrod. In th e most preferred embodiment, the matrix is extremely thin and is polymerized in sut i to form a biodegradable polymeric matrix. The matrix provides controlled release of the agent over a period of time effective to prevent surgical adhesions and is biodegradable, usually within the same time frame. Examples demonstrate that the combination of the matrix and the urokinase, tPA, birudin, or ancrod is effective in preventing surgical adhesions.

Description

LOCAL DELIVERY OF FIBRINOLYSIS ENHANCING AGENTS
Background of the Invention This invention is generally in the area of delivery of agents for prevention of surgical adhesions, and specifically involves use of locally formed topical gel systems for controlled delivery of fibrinolysis enhancement agents, especially urokinase, for improved prevention of adhesions.
Adhesions in Surgery The formation of adhesions, or scar tissue bridges, following surgery, remains a serious complication of many surgical procedures. These include pelvic, abdominal, spinal, tendon, ophthalmic, urinary, thoracic and other procedures. Adhesion formation is believed to occur through a series of events, one of which is the formation of fibrin bridges from a serosanguinous exudate occurring after surgery. The organs are first connected by thin fibrin bridges. Over time, these bridges become populated by cells, which may secrete collagen and otherwise stabilize the bridge. It has been observed that the level of cellular secretion of plasminogen activators, which normally cause the breakdown of fibrin by activating the enzyme plasminogen, can be decreased following injury to tissues. Thus, prevention of the stabilization of such fibrin bridges, and in particular enhancement of natural processes which can remove such bridges before their stabilization into adhesions, is highly desirable in the prevention of adhesions.
Cellular secretion of plasminogen activators, e.g., by the mesothelial cells that line the peritoneum, has been demonstrated by Vipond, et al., "Peritoneal Fibrinolytic Activity and Intra-abdominal Adhesions." The Lancet, 335:1120-1122 (1990), to be reduced following injury, leading to the lack of resorption of the fibrin bridges prior to maturation into a scar. In the peritoneal cavity, the particular fibrinolytic reduced in secretion was demonstrated to be tPA and not uPA.
Barrier Methods in Adhesion Prevention:
Several physical barrier methods have been utilized in the prevention of postoperative adhesions. These include sheets of oxidized regenerated cellulose (U.S. Patent No. 5,007,916 to Linsky and Cunningham; U.S. Patent No. 5,134,229 to Saferstein, et al.) sheets of expanded polytetrafluoroethylene (Buyers, et al., "Reduction of Postoperative Pelvic Adhesions in the Rabbit with Gore-Texi.~,, Surgical Membrane." Fertility and Sterility, 49:1066-1070 (1988)), thermoreversible hydrogels (U.S. Patent No. 5,126,141 to Henry), and photopolymerized, resorbable hydrogels (U.S Patent No. 5,410,016, entitled "Photopolymerizable Biodegradable Hydrogels as Tissue Contacting Materials and Controlled-Release Carriers" filed March 1, 1993 by Hubbell, et al., to which this application claims priority).
With the exception of the method of Hubbell, et al., these ~iPti~f~OED SHfEF

w0 9SI13717 CA 0 217 8 4 8 7 19 9 9 -11- 0 5 PCT/OS94114Z13 methods have ranged is usefulness, but in no case do the methods eliminate the formation of postoperative adhesions.
Use of Fibrinolytic enzymes in Prevention of Adhesions.
Various fibrinolysis enhancing agents have been used in attempts to prevent adhesions. Because of their availability and biological suitability, streptokinase (SK), urokinase (UK; also known as urokinase plasminogen activator, uPA), tissue plasminogen activator (tPA), and a modified recombinant tPA (Fb-Fb-CF) have been most widely tested. These agents all work by activation of the enzyme plasminogen, causing it to lyse fibrin.
Other substances investigated for removal or prevention of fibrin strands have included proteolytic enzymes, drugs, and clotting inhibitors such as heparin, which tend to prevent deposition of additional fibrin, referred to herein as "fibrinolysis enhancing agents".
Fibrinolytic enzymes have been used in the prevention of postoperative adhesions, as reviewed by Dune, "Tissue-type Plasminogen Activator and Adhesion Prevention." Prog. Clin. B~ol. Res.
38:213-220 (1993), and "Adhesion, Adhesiolysis and Plasminogen Activators." Assisted Human Reproductive Technology, 13:130-137 (1991).
Tissue Plasminogen Activator (tPA).
Tissue Plasminogen Activator (tPA) has been shown to be of use in the prevention of postoperative adhesions, when delivered by minipump infusion, intraperitoneal injection, and topically.
Sheffield (Exp.eye res. 52(6):73-741, 1991) describes the topical administration of tPA preferably by injection, but possibly in a phospholipid carrier, a salve or ointment, a polysaccharide composition, a thermoplastic polymeric gel or a hydrogel such as a polyoxyethylene-polyoxypropylene block copolymer,.

WO 95J13747 PCT/089dI14I13 which releases over a period of between three hours up to seven days. Wiseman, et al. (General Surgical Research vol. 53 (4):36-368,1992), describes the addition of tPA, a tPA analog, Fb-Fb-Cf, and SK, alone or in combination with an absorbable sheet of oxidized regenerated cellulose, Interceed TC7TM
absorbable adhesion barrier from Ethicon, Inc., Somerville, NJ.
Streptokinase (SK) .
Streptokinase (SK), the earliest plasminogen activator to become widely available, has been shown by some investigators to be effective in preventing adhesions and ineffective by others. SK has been shown to be effective in the prevention of postoperative adhesions when delivered by peritoneal injection (Meier, et al. "First Clinical Results of Intraoperative Application of Streptokinase-Streptodornase in Children." Langenbecks Archiv.
fur Chirurgie, 366:191-193 (1985), Treutner, et al.
"Postoperative, intraabdominelle Adh~sionen-Ein neues standardisiertes and objektiviertiea Tiermodell and Testung von Substanzen zur Adhasionsprophylaxe.*" Lagenbecks Archiv. fir Chirurgie, 374:99-104 (1989)) and ineffective in other studies when similarly delivered (Verreet, et al. "Preventing Recurrent Postoperative Adhesions:
An Experimental Study in Rats." Eur. Surg. Rea., x1:267-273 (1989)). When delivered as a continuous infusion, it was not effective (Sheffield). When delivered from a degradable polymer matrix, it was somewhat effective (Wiseman, et al.).
Reports on side-effects of tPA differ widely, from none, at effective doses (Menzies and Ellis, or Sheffield), to severe at effective doses (Wiseman, et al . ) .
Fb-Fb-CF.
Fb-Fb-CF is a tPA analog (Phillips, et al., "The Effects of a New Tissue Plasminogen Activator WO 95!15747 PCTlUS94/14213 Analogue, Fb-Fb-CF, on Cerebral Reperfusion in a Rabbit Embolic Stroke Model." Annals of Neurology, 23:281-285 (1989)) and was effective in the prevention of postoperative adhesions when released 5 from a degradable polymer matrix (Wiseman, et al.).
Genetic engineering~is being applied to generate additional forms of plasminogen activators, plasmin, plasminogen, and other fibrinolytic agents, for example, as described in U.S. Patent No.
5,223,408, 5,185,259, and 5,094,953.
Urokinase Plasminogen Activator (uPA).
The majority of studies with Urokinase Plasminogen Activator (uPA) have not demonstrated usefulness in preventing surgical adhesions. uPA
has been investigated in the prevention of adhesions, as reported by Dunn (1991). An initial study with uPA in several species was not successful. However, a second study by Gervin, et al., "A Cause of Postoperative Adhesions" Am. J.
Surg. 125:80-88 (1973), did show efficacy. Dogs were treated with an intraperitoneal injection of large amounts of urokinase at the time of surgery.
At dosages of 20,000 U/kg, there was allegedly a significant decrease in the formation of ilea!
adhesions.
However, no subsequent studies have demonstrated efficacy. For example, as reported by Rivkind, et al. "Urokinase Does Not Prevent Abdominal Adhesion Formation in Rats." Eur. Surg.
Res.,17:254-258 (1985), who studied the administration of urokinase in dosages between 5,000 and 100,000 U/kg administered intravenously, intraperitoneally, and intragastrically immediately poatoperatively and at 48-72 hours post surgery.
Moreover, uPA released continuously with minipumpa at the site of injury did not reduce adhesions, as reported by Sheffield, et al. Another study in R'O 95115747 PCTlUS941i4213 rabbits claimed to show efficacy with abrasion injury to the uterine horns, following administration of 10,000 U/kg urokinase either , intraperitoneally or intravenously at the time of surgery and at 24 and 48 hours after surgery, both , when blood was added to the peritoneal cavity and without added blood (injury in both cases), Birkenfeld and Schenker "The Effect of Urokinase in the Prevention of Intraperitoneal Adhesions; Role of Blood in Their Formation." Annalea Chirurgiae et Gynaecologiae, 72:246-249 (1983). Notwithstanding the assertion of efficacy, reference to Table 2 of Birkenfeld and Schenker, shows that comparison between the group !'Serosal abrading of the right-uterine horn - No treatment" and the group "Serosal abrading of the right uterine horn - Urokinase treatment" demonstrates a difference with a p value of only approximately 0.2. By contrast, the group "Serosal-abrading + application of blood on right uterine horn - No treatment" differs from the group "Serosal abrading + application of blood on right uterine horn - Urokinase treatment" with a p value <0.02, both by Chi-squared teat. Thus, the treatment with urokinase was effective only with the addition of an unnatural amount of blood.
In-summary, uPA has been shown to be effective in the prevention of adhesions in only one study, and it has not been possible to reproduce this observation using normal conditions and normal standards for statistical difference. Efficacy with other fibrinolytic agents has also been mixed, with the best results for tPA being obtained only with repeated or continuous infusion of high dosages,-and -essentially no efficacy being observed with streptokinase.
The literature in this field is highly equivocal about the utility of any of these agents 2~ 78+87 , as reliable, clinically effective means for prevention of surgical adhesions.
EP-A 0 472 011 discloses a composition that includes a protease and a pharmaceutically acceptable carrier for treating wounds. AU-A-38229/93 discloses the use of inhibitors of plasminogen activators for treatment of inflammations and wounds.
h is therefore an object of the present invention to provide a reliable means for locally preventing surgical adhesions using fibrinolytic agents. It is a further object of the present invention to provide a means for preventing surgical adhesions which is minimally invasive, biodegradable over the same period of time as healing occurs, and simply to use. It is still another object of the present invention to provide a means for preventing surgical adhesions which uses low dosages of ftbrinolytic aeents.
Summary of the Invention Described herein is a method of preventing adhesions by topical administration of f'brinoIysis enhancing agents. In particular, described herein is a method using a local, degradable polymeric release system for the prevention of surgical ad-hesion using a topically applied polymeric matrix for delivery of a fibrinolytic agent, preferably urokinase, streptokinase, hirudin or ancrod. In the most preferred embodiment, the matrix is extremely thin and is polymerized in situ to form a biodegrad-able polymeric matrix. The matrix provides controlled release of the agent over a period of time effective to prevent surgical adhesions and is biodegradable, usually within the same time frame. In contrast to prior studies with urolcinase administered systemically, the combination of the matrix and the uroliittase is effective in preventing surgical adhesions, as shown by comparative examples. The examples also demonstrate efficacy AMENDED SHEET

VJO 95115747 pCT/US94/14213 for tissue plasminogen activator (tPA) administered via a controlled release, biodegradable system using low dosages.
Detailed Description of the Invention Disclosed herein are delivery systems for the effective and reliable delivery of fibrinolysis-enhancing agents (FEAS) for the prevention of surgical adhesions.
Fibriaolysie-enhancing agents.
FEAs include agents such as urokinase, streptokinase, tissue plasminogen activator, hirudin, snake venom components such as ancrod, and genetically engineered variants thereof, all of which operate by binding to plasminogen and causing it to become proteolytically active against fibrin.
FEAs also include thrombolytic agents; proteins directly attacking fibrin, such as proteases, including plasmin or plasminogen; molecules which inactivate a class of proteins which themselves inhibit plasminogen (anti-plasmin inhibitors);
heparin and other fibrin-deposition Mockers; and other agents having the effect of locally enhancing the dissolution of fibrin, or preventing its deposition.
Am effective amount of one or more of these materials are administered topically to an area where adhesions are to be prevented. As used herein, "adhesions~~ includes surgical or postsurgical adhesions, postinfection adhesions, strictures, scar tissue formation, and related processes. Adhesions can occur in pelvic, abdominal, spinal, tendon, ophthalmic, urinary, thoracic and other procedures.
In the preferred embodiment, the agent is urokinase or tPA, most preferably urokinase.

Dosages of FEAs are determined based on routine optimization from dosages currently in use as delivered systemically or by injection, or dosages that have been reported in the literature.
In rats, tPA may be used-in doses of 0.03 mg per animal up to doses of 12 mg per animal, in a treatment series; present evidence suggests a preferred range, when administered in a polymeric material, in the range of 1 to 2 mg per dose.
Appropriate dosages for humans can be obtained by extrapolation; the above rat dose corresponds to the therapeutic amount normally administered for systemic fibrinolysis after heart attack, and thus may be near the upper end of the range. Urokinase likewise has proved effective at doses suitable for systemic application, about 45,000 International Units in the rat, and lower concentrations may be effective. The required range of streptokinase is likely to be higher than the equivalent of 15,000 IU
per rat.
Means for topical administratiaa.
Although a number of methods are known for topical delivery, those which have been discovered to be effective yield controlled release of incorporated FEAS over a period of time required to prevent adhesions from forming. The most effective and requiring least patient compliance are those consisting of a biocompatible biodegradable polymeric matrix which can be applied to the site where adhesions are to be prevented at the time of surgery.
The polymeric matrices should meet the following criteria:
The material should be biocompatible and biodegrade, either by hydrolysis or enzymatic cleavage, over a period of between a day and thirty days, for example, although longer degradation times R'O 95115747 ~ PCTIUS94/14213 may be desirable in some cases, as determined by the nature of the adhesions to be prevented and their location. In the abdomen, the preferred period is between one day and one month, preferably two days 5 to two weeks, and most preferably, four days to one week.
The material should be in a form which is either (1) applied as a liquid or a gel, which is then polymerized or crosslinked in situ to provide 10 additional physical integrity or binding to the tissue to be affected, or (2) be conformable and securable to the site where adhesion is to be prevented. These materials are distinguished from materials which are preformed before delivery to the site, such as Teflon~ membranes, oxidized cellulose cloth, or gelatin sponges. The ungelled polymer may be delivered by any suitable means; the preferred means will vary with the particular situation in which adhesions are to be prevented, or fibrinolysis is to be enhanced. If the operation is conducted with laparoscope or trochar, then a preferred method is to spray an ungelled barrier material, preferably containing the fibrinolysis enhancing agent, onto the site of damage; and then to gel the barrier-with light, or by other means suitable to the particular barrier system. If the operation involves an open wound, then it may be advantageous to apply the- barrier by syringe orb similar means. If the barrier is to be emplaced inside a hollow vessel or lumen, then use of a catheter for delivery may be moat appropriate.
The thickness of the polymeric material will be determined by several factors, including the amount of FEA to be delivered, the time during which the FEA is to be delivered, the thickness of the barrier required for mechanical resistance, and -WO 95!15747 PCTIUS94II42I3 similar factors. Within these limits, thinner barriers are preferred, since they will erode more rapidly, and often will be quicker to set, and are less likely to obstruct other organs or biological processes, such as blood flow. In general, the thickness contemplated are in the range of 1 to 500 microns. Ranges of 10 to 100 micron are preferred, and ranges of 20 to 50 microns are most preferred.
However, thicker barriers are contemplated if required to contain or delivery the appropriate amount of FEA.
The erodability of the material is subject to control through selection of.the composition and the stability of its degradable links. The requirements for stability of the material are that it persist long enough to achieve the desired effect, and that it then erode completely in as short a period as feasible. For the prevention of adhesions in the abdomen, current data suggest that the barrier, and the delivery of the FEA, should continue for at least one day, preferably for two or more, and most preferably far at least four days. In spinal disk surgery, fragmentary evidence suggests that periods of 1 to 2 weeks may be preferable, to prevent post-operative adhesions. It is preferable that the barrier complete its disintegration in less than a month, and more preferably in a shorter period such as two weeks, provided that this period is at least as long as the time required for the therapeutic presence of the FEA.
The material must also not be reactive with the incorporated FEA so as to inhibit their effectiveness, but must serve to control the rate of release of FEA at the size, preferably by diffusions! restrictions. Since the time for healing is generally in the range of two weeks, the preferred timse for release is in the range of wo ~ns~s~ pcriossvmu3 between one day and up to thirty days. Control of diffusion is accomplished by design of pore sizes, relative affinity of the barrier matrix for the FEA, partial insolubility of the FEA, and physical size and shape, of the deposited barrier. For short treatment times, of the order of a few days, the simple anticonvective effect of a porous hydrophilic matrix may give sufficient control.
The material may also function to prevent adhesion through properties of the polymer per se, which is cumulative to the action of the FEA. Such materials are described in U.S. Patent No.
5,410,016 entitled "Photopolymerizable Biodegradable Hydrogels as Tissue Contacting Materials and Controlled-Release Carriers" filed March 1, 1993 by Jeffrey A. Hubbell, Chandrashekhar P. Pathak, Amaxpreet S. Sawhney, Neil P. Desai, and Jennifer L. Hill, Briefly, polymers are formed from biocompatible, biodegradable, polymerizable and at least substantially water soluble macromers, having at least one water soluble region, at least one region which is biodegradable, usually by hydrolysis, and at least two free radical-polymerizable regions. The regions can, in some embodiments, be both water soluble and biodegradable. The macromers are polymerized by exposure of the polymerizable regions to free radicals generated, for example, by photosensitive chemicals and dyes.
An important aspect of the macromers are that the polymerizable regions are separated by at least one degradable region to facilitate uniform degradation is vivo. There are several variations of these polymers. For example, the polymerizable regions can be attached directly to degradable extensions, or indirectly via water soluble W095115747 ~ ~ ~ ~ (~ ~ 7 PCT/ITS94/I42I3 nondegradable sections so long as the polymerizable regions are separated by a degradable section. For example, if the macromer contains a simple water soluble region coupled to a degradable region, one polymerizable region may be attached to the water soluble region and the other attached to the degradable extension or region. In another embodiment, the water soluble region forma the central core of the macromer and has at least two degradable regions attached to the core. At least two polymerizable regions are attached to the degradable regions so that, upon degradation, the polymerizable regions, particularly in the polymerized gel form, are separated. Conversely, if the central core of the macromer is formed by a degradable region, at least two water soluble regions can be attached to the core and polymerizable regions attached to each water soluble region. The net result will be the same after gel formation and exposure to in vivo degradation conditions. In still another embodiment, the macromer has a water soluble backbone region and a degradable region affixed to the macromer backbone.
At least two polymerizable regions, are attached to the degradable regions, so that they are separated upon degradation, resulting in gel product dissolution. In a further embodiment, the macromer backbone is formed of a nondegradable backbone having water soluble regions as branches or grafts attached to the degradable backbone. Two or more polymerizable regions are attached to the water soluble branches or grafts. In another variation, the backbone may be star shaped, which may include a water soluble region, a biodegradable region or a water soluble region which is also biodegradable.
In this general embodiment, W0 9S/15747 . PCT/US94/14213 the star region contains either water soluble or biodegradable branches or grafts with polymerizable regions attached thereto. Again, the photopolymerizable regions must be separated at some point by a degradable region.
Examples of these macromers are PEG-oligolactyl-multiacrylates. The choice of appropriate end caps permits rapid photopolymerization and gelation; acrylates are selected because they can be polymerized using several initiating systems, e.g., an eosin dye, by brief exposure to ultraviolet or visible light. The poly(ethyleneglycol) or PEG central structural unit (core) is selected-on the basis of its high hydrophilicity and water solubility, accompanied by excellent biocompatibility. A short poly a-hydroxy acid), such as polyglycolic acid, is selected as a preferred chain extender because it rapidly degrades by hydrolysis of the eater linkage into glycolic acid, a harmless metabolite. Although highly crystalline polyglycolic acid is insoluble in water and most common organic solvents, the entire macromer is water-soluble and can be rapidly gelled into a biodegradable network while in contact with aqueous tissue fluids. Such networks canbe used to entrap and homogeneously disperse water-soluble drugs and enzymes, such as the FEAs, and to deliver them at a controlled rate. Other preferred chain.
extenders are polylactic acid;-polycaprolactone, polyorthoestera, and polyanhydrides. Polypeptides and polysaccharides may also be used.
These materials are particularly useful for controlled drug delivery, especially of hydophilic materials such as most of-the FEAS, since the water soluble regions of the polymer-enable access of water to the materials entrapped within the polymer.
Moreover, it is possible to polymerize the macromer WO 95/15747 2 ~ 7 $ Q ~ l PGT/US94/14213 containing the material to be entrapped without exposing the material to organic solvents. Release may occur by diffusion of the material from the polymer prior to degradation and/or by diffusion of 5 the material from the polymer as it degrades, depending upon the characteristic pore sizes within the polymer, which is controlled by the molecular weight between crosalinks and the crosalink density.
Deactivation of the entrapped material is reduced 10 due to the immobilizing and protective effect of the gel and catastrophic burst effects associated with other controlled-release systems are avoided. When the entrapped material is an enzyme, the enzyme can be exposed to substrate while the enzyme is 15 entrapped, provided the gel proportions are chosen to allow the substrate to permeate the gel.
Degradation of the polymer facilitates eventual controlled release of free macromolecules in vivo by gradual hydrolysis of the terminal ester linkages.
An advantage of these macromers are that they can be polymerized rapidly in an aqueous surrounding. Precisely conforming, semi-permeable, biodegradable films or membranes can thus be formed on tissue in situ. In a particularly preferred embodiment, the macromera are applied to tissue having bound thereto a photoinitiator, and polymerized to form ultrathin coatings. This is especially useful in forming tissue barriers during surgery which thereby prevent adhesions from forming.
Examples in this application demonstrate the use of these macromers and polymers for the prevention of postoperative surgical adhesions in rat cecum and rabbit uterine horn models. The polymers show excellent biocompatibility, as demonstrated by minimal fibrous overgrowth on R'O 95/15747 PCT/US94/14213 implanted samples. Hydrogels for the models were gelled in situ from water-soluble precursors by brief exposure to long wavelength ultraviolet (LWUV) light, resulting in formation of an interpenetrating network of the hydrogel with the mucous, serous, or , serosanguinous layer coating the tissue. The degradable hydrogel was very effective, both by itself and in combination with tPA, in preventing adhesions.
These materials are distinguished from materials such as oxidized regenerated cellulose cloth, or perfluoroalkylene membranes, by their ability to be delivered by syringe, catheter, spray or solution, which enables their-delivery to sites of tissue injury through less invasive techniques.
Moreover, being biodegradable or bioerodable, they do not require removal. They are distinguished from liposomes, from polymeric adjuvants such as dextran, and from non-firmly gelled polymers such as hyaluronic acid, by their ability to be precisely localized at the site or sites where adhesion prevention is required. This minimizes the dose of FEA required to produce the desired effect.
Alternative polymeric materials include materials which are settable by other mechanisms.
These include polymers which set on,warming, such as poloxamera (Pluronic'~ or Butyronic~ detergents;
block copolymers of polyethylene oxide and polypropylene oxide or polybutylene oxide), or hydroxypropyl methyl cellulose; polymers which gel on cooling, such as gelatin; polymers which can gel on exposure to physiological ions (calcium), such as alginate or chitosan; polymers which gel by other mechanisms, such as redox-changes; provided in all cases that the settable polymers are also ' sufficiently inert and erodible.

~ 2178487 The present invention will be further understood by reference to the following non-limiting examples.
Example 1: Prevention of adhesions by release of fibrinolyrtics from hydrogel polymeric matrices.
Efficacy testing was done in the rat, with injury to the uterine horn by eIectrocautery as described by U.S Patent No. 5,410,OI6, to produce a devascularization and serosal injury. Animals were sacrificed seven days following injury and treatment, and adhesions scored. The incisions were reopened and adhesions were scored for extent and tenaciy. Extent of adhesion formation was evaluated by measuring the length of the uterine hom that formed adhesions with itself or with the peritoneal wall or other organs. Tenacit3~ of adhesion was classified as either fMmy or fibrous.
Filmy adhesions were usually transparent, less strong, and could be freed by hand. The fibrous adhesions were dense, whitish, and usually required sharp instrument dissection to be freed. In cases where only a single filmy adhesion band was evident, a score of 5 %a was assigned.
Typical samples of the horn were excised for histology and were feed in a IO% rieiitral buffered formalin solution. Paraffin sections of the samples were Brained using hematoxylin and eosin.
The adhesion score is the % of affected area occupied by the adhesions, with grading of each as being filmy or fibrous.
Sexually mature female rates were-prepared for surgery. .A midIine incision was made in the lower abdominal region under Rompun, Ketamine, and Acepromazine anesthesia. The uterine horns were located and the vasculature to both horns was systematically cauterized to induce an ischemic injury. .~.fter cauterization, macromer solutions Hv4EP~D'~D SHEEN

W095/15747 CA 02178487 1999-11-OS p~/pgg4/14Z13 (0.5 ml) were applied along the horn and allowed to coat the surface where the cauterization injury had been induced. After uniform application of the macromer solution was complete, the horns were exposed to a LWUV lamp for 1 min to induce gelation. The procedure was repeated on the reverse side of the horns. The incisions were then closed using a continuous 2-0 VicrylTM (Ethicon) suture for the musculoperitoneal layer and a 0 VicrylTM
(Ethicon) suture for the cutaneous layer. No prophylactic antibiotics were administered. The ischemic injury was made as described and the incision was closed without the application of the precursor; all techniques were identical between the treatment group and the control group.
Controls were used where the same animal model was subjected to surgery without application of the macromer; all surgical techniques were identical between the treatment group and the historical controls.
Animals were treated with a photopolymerized hydrogel barrier, with or without fibrinolytic agents included in the barrier. A viscous sterile 15% solution in phosphate buffered saline (8.0 g/1 NaCl, 0.201 g/1 KC1, 0.611 g/1 Na2FiP0" 0.191 g/1 KHPO" pH 7-4) of polyethylene glycol (M. W. 18,500) which has been chain extended on both ends with a short polyglycolide repeat unit (average number of glycolidyl residues: 10 on-each end) and which has been subsequently terminated with an acrylate group was prepared. Initiator needed for the crosslinking reaction, 2,2-dimethoxy-2-phenyl acetophenone, was added to the macromer solution to achieve an initiator concentration of 900 ppm. A 30 second exposure to a long wave W lamp (Blak RayTM) is sufficient to cause polymerization. The fibrinolytic proteins were mixed in the macromer R'O 95115747 PCTlUS94/14213 solutions in the concentrations shown in the footnotes to Table 1.
SK was obtained from Astra Pharmaceutical Products, Westborough, PA; tPA from Genentech, South San Francisco; CA, and urokinase from Abbott Laboratories, North Chicago, IL. A base dose of 3 tPA mg/ml (net 1,740,000 units) was selected. The equivalent doses for SK (45,000 units) and uPA
(15,000 units) were calculated based on the relative doses used to treat myocardial infarction, the primary use of the drugs: dose of agent = (3 mg/ml) x (dose of agent used to treat myocardial infarction)/(dose of tPA used to treat myocardial infarction). Each experimental group consisted of 7 rats.
Results and statistical analyses are shown in Table 1.
Table 1: Treatmeat of Adhesions by Release of Fibriaolytics from Hydrogel Polymer Matrices.
Group n Mean Std. Mean Extents, % Dev. , '% Grade Control 7 72 15 1.7 Gel * 7 22 10 1.1 Gel + SKt 7 45 25 1.7 Gel + 7 6 6 1.0 uPAt Gel + 7 4 3 1.0 tPAt * 8KL5, 15% macromer in saline.
3D t Concentrations of proteins in gel precursor were as follows: SK, 1.2 mg/ml; uPA, 1.8 mg/ml, tPA, 3 mg/ml; 1.5 ml per animal, as described above.
~ Each group is significantly different than the others by Kruskal-Wallis at c confidence level of 95%, except the Gel+uPA
and the Gel+tPA groups.

W O 95115747 ~ PCTICTS94114213 2~1 '7$4$7 uPA in solution was ineffective, even over a four-day course of intraperitoneal injection.
Solutions of tPA were only somewhat effective, reducing adhesions by about one-third of the 5 control. The barrier gel alone, without FEA, , reduced adhesions by about 75% of the control.
However, the barrier gel and uPA and tPA reduce adhesions significantly, to 4-6%, or by about a factor of five below the gel only control. The 10 barrier gel alone is more effective than any of the FEAs tested alone, but the combination is strikingly more effective than either.
Example 2: Comparison with treatment of adhesions by daily injections with fibrinolytic 15 agents.
To understand the importance of the degradable polymer controlled release system in the efficacy of uPA, the same total dose of the various fibrinolytic drugs was delivered in the same model 20 over a four day period, with intraperitoneal injection of one-fourth of the total dose daily.
Solutions of sterile drugs were made up in sterile HEPES-buffered saline (pH 7.4, 10 mill) at the following concentrations: tPA, 3 mg/ml; uPA, 1.8 mg/ml; SK, 1.2 mg/ml. The adhesion model was performed exactly as described in Example 1. A dose of 0.375 ml was given intraperitoneally immediately after injury prior to closure of the peritoneal cavity. An equal dose was administered on each of the next three days by intraperitoneal injection.
Thus, the total dose of each drug was identical to that in the study with the degradable polymeric release system. Control animals received an equal volume of HEPES-buffered saline. Each group consisted of seven animals (one animal died in the SK treatment group in the postoperative period, due to anesthesia complications). Adhesions were scored WO 95115747 , PCTlUS941142l3 on the seventh postoperative day, as described in Example 1.
Results and statistical analyses are reported in Table 2.
Table 2: Treatment of Adhesions by Daily Injection of Fibrinolytics.
Group n Mean Std. Mean Extents,% Dev., % Grade Control 7 77 16 1.6 SKt 6 83 13 1.7 uPAt 7 78 18 1.7 II tPAt I 7 49 21 I.5 t Doses of Proteins were as follows: SK, 1.2 mg/ml, 0.375 ml/doae, 1 dose per day, days 0, I, 2 and 3; uPA, 1.8 mg/m1,0.375 ml/doae, 1 dose per day, days 0, 1 and 3; tPA, 3 mg/ml, 0.375 ml/doae, 1 dose per day, days 0, 1. 2 and 3.
~ The tPA group is significantly different than the others by Kruskal-Nlallis at a confidence level of 95%; no other group is significantly different.
The daily injection of tPA was somewhat effective, although much less so than the delivery of tPA from the degradable polymeric system. The daily injection of uPA was not effective. Thus, one may conclude that the drugs alone were not responsible for the efficacy observed in Table 1. Rather, the drug formulation in a controlled release formulation was effective.
The studies in Examples 1 and 2 were performed in a blinded and randomized manner. The surgeon performing the injury was blinded to the grouping of the animals. The surgeon administering the gels was blinded as to the content of the gels, and the surgeon administering the injections was blinded as to the content of the injections. The surgeon performing the scoring of adhesions was blinded as to the grouping of the animals.
8xample 3. Use of Aacrod to Prevent Adhesioas.
Ancrod is an enzyme from the venom of the Malaysian Pit Viper, Agkiatrodon Rhodoatoma (c.f.- , Merck Index, 10th ed. #655). Known as an anticoagulant, it acts to prevent clotting by the hydrolysis of fibrinogen, thereby preventing the formation of fibrin strands. Ancrod was obtained from Sigma Chemicals.
Using procedures as described in Example 1, Ancrod was administered by various routes and the percent of adhesion was scored. Ten animals were used for each study. Groups were divided into saline controls; one IP injection of 10 units/kg body weight of ancrod on the day of operation; and three injections.
MATERIALS AND METHODS:
Adhesion Model Female Sprague Dawley rata weighing 240 to 260 g were used for the study. At the time of induction of injury, the animals were anesthetized with 40 mg/kg of pentobarbital injectedintraperitoneally. The abdomen was shaved, cleaned with Betadineaolutionand opened by a lower midline longitudinal incision. The uterine horns were exposed and the vasculature supplying the uterine horns was systematically cauterized with a bipolar electrocautery. Additionally, two cautery injuries, each approximately 1 mm in diameter, were made 2.5 cm apart on the antimesenteric aerosa of each uterine horn. The uterine horns were replaced in the abdominal cavity and the abdominal wall-was I closed in two layers, the musculi-peritoneal with continuous 4-0 Vicryl~ sutures (Ethicon) and the skin with 9 mm wound clips.

pGTlU894/142I3 On the seventh postoperative day, the rats were sacrificed by carbon dioxide asphyxiation. The abdomen was opened and the uterine horns were exposed. The total length of each uterine horn and , 5 the length of each horn that was involved in adhesions were measured with a ruler. The fraction of the total uterine horn length involved in adhesions was calculated for each animal after addition of the values for both uterine horns. The quality of the adhesion, regardless of its extent, was also scored as grade 1 or 2. Grade 1 was assigned in the case of filmy adhesions, which could be easily separated by hand; and grade 2 in the case of dense adhesions which could not be separated without dissection by sharp instruments.

Results on the extent of adhesions were analyzed by the Kruskall-Wallis test to determine trends in treatment groups, and comparison between groups were made by the Marin-Whitney U-test. Results on the grade of adhesions were compared between groups by the Chi-squared test.

Administration of Ancrod by Pre- through nRtnnPrafi~rc T"r,-a<rc..~.;.. r..,..
-.....-.... ~ 1 I La 1311 To establish a dose-response relationship, ancrod was delivered by IV injections 5 days pre-through 3 days postoperatively. Animals were randomly divided into four experimental groups. The first group received saline, and other three groups received 5, 10 or 20 units/kg/day of ancrod administered in two equally divided injections per day for 5 pre-through 3 postoperative days. Ancrod was obtained as alyophilized powder in 10 or 50 units vial (Sigma Chemicals, St. Louis, Mo.). The powder form of ancrod was reconstituted, under sterile conditions, to the desired strength by addition of R'O 95/15747 PCT/US94114213 water for injection. Each injection was of equal volume of 0.25m1/rat.
One-half day following the tenth injection, the , uterine horn injury was performed as described above.
Immediately following the operation the animals received the eleventh injection, and the injections were continued through the eighth day, that being three days poatoperatively. On the seventh postoperative day, the animals were sacrificed and the adhesions were measured and graded as described above. Both the injury and the measurement and -scoring were performed such that the surgeon (S.M.C.) was blind to the animal grouping.
Administration of°Ancrod by Pre or Postoy~erative Intraberitoneal Injection Two studies were performed to determine whether IP delivery represents an effective route of administration, and further to determine whether defibrinogenation before or after the electrocautery injury was more effective.
In the model with preoperative injection of ancrod, animals were randomly divided into 4 experimentalgroups. The first group received IP
saline 5 days prior to the surgery, twice daily. -The other three groups received 5, 10 or 20 units/kg/-day of ancrod by IP injection a.n two equally divided doses. The last injection was given on the morning of surgery. Following the final injection, the animals were operated as described above. On the seventh postoperative day,the animals were sacrificed and the adhesions were measured and graded as described above. Both the injury and the measurement and grading were performed blindly to the grouping of the animals.
In the model with preoperative injection of ancrod, animals were randomly divided into 4 experimental groups. The first group received saline R'O 95/15747 PCT/ITS94/14213 and the other three groups received 5, 10 or 20 units/kg/day of ancrod by IP injection in two equally divided doses for three days poatoperatively. The animals were operated as described above and the 5 first dose was administered within 2 hours of surgery. On the seventh postoperative day, the animals were sacrificed and the adhesions were measured and graded as described above. Both the injury and the measurement and grading were performed 10 blindly to the grouping of the animals.

Administration of Ancrbd by Local Release from a 10% Hvdrooei An experiment was performed to determine whether it was possible to effectively deliver ancrod to the 15 injured uterine horns locally by slow release from a degradable hydrogel. This hydrogel was photopolymerized in situ from an aqueous precursor solution. A 10% w/v prepolymer solution was prepared immediately before use, under sterile conditions, by 20 dissolving 1 g of prepolymer (supplied as dry ower p from Focal Inc., Lexington, MA) in lOmi of water for injection. The long-wavelength ultraviolet photoinitiator 2,2-dimethoxy, 2-phenyl acetophenone (Aldrich Chemical Co., Milwaukee, WI) was dissolved 25 in N-vinyl pyrolidinone (Aldrich) at a concentration of-600 mg/ml, and 1.5 ~1 of this initiator solution was added to lmi of the prepolymer solution to achieve a final concentration of 900 parts per million initiator in this preclinical research formulation. The prepolymer solution was sterilized by passage through a 0.2 ~ filter with a 0.8 ~Cm prefilter (Millipore, Bedford, MA). The prepolyer solution-was tested for gelation by illuminating one drop of the solution with a hand-held long-wavelength (365 nm) ultraviolet lamp at an intensity of approximately 70 mW/cma (Black Ray, UVP, Inc., San Gabriel, CA). Conversion to a gel occurred within R'O 95115747 ' PCTIUS94114213 s approximately 7-8 sec. Fifty units of anerod was dissolved in 0.5 ml of water for injection and 0.1 ml was added to 4 ml of -prepolymer solution to deliver , units of ancrod/kg over approximately three days 5 time period, 1 ml prepolymer being administered to , each animal. This mixture was again tested for gelation.
The effect of ancrod delivered locally via the hydrogel was compared with delivery by IP injections.
10 Animals were randomly divided into 5 groups. The first group was control and received no treatment, the second group received only the 10% gel, and third group received 10 units/kg of ancrod in the 10% gel.
The fourth group received 10 units/kg of ancrod as one postoperative IP injection and the fifth group received 10 units/kg of ancrod over three days postoperatively as 3 equally divided IP injections.
Uterine born injury was performed as described above.
In the groups receiving gel with or without ancrod, 0.5 ml of the prepolymer solution was applied to each born and was gelled by exposure to long-wavelength ultraviolet light for 20 seconds each at an intensity of approximately 70 in W/cm2on the ventral and dorsal side. The animals were sacrificed on the seventh-postoperative day and adhesions were measured and graded as described above. Both the injury and the measurement and grading were performed blindly to the grouping of the animals.
Dose Response Locally Delivery of Ancrod From a 10% hvdroael A study was performed with higher doses of ancrod, locally delivered through a 10% hydrogel_ Animals were randomly divided into 5 experimental groups. The first group was control and-received no treatment. The second group received the 10% hydrogel only, as described above. The other three groups received 25, 50 or 100 units/kg of -ancrod locally WO 95!15747 ~ ! 7 $ 4 8 7 p~~s9'ut42I3 delivered through the 10% hydrogel. Fifty units of ancrod was dissolved in 0.5 ml of water for injection , under sterile condition and 0.25, 0.5 or 1 ml of this solution was added to 4 ml of prepolymer solution to deliver 25, 50 or 100 units/kg of ancrod over approximately 3 days time period. The initiator was added, the solution was sterilized and the hydrogel was applied as described above. On the seventh postoperative day, the animals were sacrificed and adhesions were measured and graded as described above. Both the injury and the measurement and grading were performed blindly to the grouping of the animals.
Administration by Local Release From a 15%
- Hydrocrel A study was performed to provide sustained release of ancrod from a 15% hydrogel, which is expected to release the drug somewhat more slowly.
Fifty units of ancrod was dissolved in 0.5 ml of water for injection and 0.5 ml of this solution was added to 4 ml of prepolymer solution to deliver a dose of 50 units/kg of ancrod to each animal.
Animals were randomly divided into 3 experimental groups. The first group was control and received no treatment. The second group received the 15%
hydrogel only, and the third group received 50 units/kg of ancrod delivered through the 15%
hydrogel. The initiator was added, the solution was sterilized, and the hydrogel was applied after injury as described above. On the seventh postoperative day, the animals were sacrificed and adhesions were measured and graded as described above. Both the injury and the measurement and grading were performed blindly to the grouping of the animals.

R'O 95/15747 PCTYUS94/14213 Table 3: Administration by Pre-through Postoperative Intravenous Injection.
Number Treatment $xteat (%) Adhesion , of Animals Mean, SEMI
Grade,0-2 7 A Control 83, 3.1 2,2,2,2,2,2,1 7 B 5 units 32, 5.4 2,2,1,1,1,1,1 6 C 10 unite 31, 3.5 2,2,1,/,1,1 7 D 20 units 29, 5.6 1,1,1,1,1,1,1 *Dose refers to number of units/kg/day, injected twice daily from 5 days pre to 3 days postoperatively.
~ P < 0.001 by Kruskal-Wallis test.
Table 4: Administration by Preoperative Intraperitoneal Injection Number Treatment Lxteat (%) Adhesion of Animals Mean,SEM~ Grade,0-2 7 A Control 58, 4.6 2,2,2,2,1,1,1 7 B 5 unite 43, 6.1 2,2,1,1,1,1,1 7 C 10 units 44, 9.8 2,2,2,1,1,1,1 D 20 units 25, 8.0 1,1,1,1,1 *Doae refers to number of units/kg/day, injected twice daily from 5 days postoperatively.
~ P < 0.001 by Kruskal-Wallis test.
Table 5: Administration by Postoperative Intraperitoaeal Injection Number Treatment Extent (%) Adhesion of Animals Meaa,SEM~ Grade,0-2 6 A Control - 87, 5.2 2,2;2,2,2,2,2 7 B 5 units 39, 6.5 2,1,1,1,1,1,1 7 C 10 units 31, 2.5 1,1,1,1,1,1,1 7 D 20 units 14, 2.2 2,1,1,1,1,1,1 *Doae refers to number of units/kg/day, injected twice daily for 3 days postoperatively. .
~ P < 0.001 by Kruakal-Wallis teat.

R'O 95115747 PCTlUS94/14213 Table 6: Admi.niatration by Local Release from a 10% Hydrogel Number Treatment Extent (%) Adhesion of Animals Mean,SEM
Grada,0-2 A Control 79, 3.6 2,2,2,2,2, 2,2,2,1,1 10 B Gel + Ancrod* 14, 2.8 1,1,1,1,1, 1,1,1,1,1 IO C Gel only 25, 2.3 1,1,1,1,1, 1,1,1,1,1,1 10 D 3 Injections ip 56, 3.8 2,2,2,2,2, 2,2,1,1,1 10 E 1 Injection ip 64, 4.0 2,2,2,2,2, 1,1,1,1,1 *The total does of ancrod was 10 unita/kg.

In the Gel + Ancrod group,- the gel contained 10 units/kg total_ In the group receiving three injections, one-third of this dose was delivered in each of three daily intraperitoneal injections, postoperatively.

P<0.001 by Kruskal-Wallis test.

Table 7: Dose Raspoase Delivery from in Local a 10% Hydrogel Number Treatment Extent (%) Adhesion of Animals Meaa, SEM
Grade,0-2 7 A Control 78, 2.7 2,2,2,2,2,1,1 7 B Gel Only 29, 3.8 2,2,2,1,1,1,1 7 C 25 units in gel* 16, 2.7 2,1,1,1,1,1,1 7 D 5D units in gel 14, 1.5 1,1,1,1,1,1,1 7 E 100 unite in gel 13, 3.2 1,1,1,1,1,1,1 *Dose refers to number of units/kg total_ P< 0.01 by Kruskal-Wallis test.

WO 95/15747 ; PCT/US94114213 Table 8: Administration by >;ocal Raleasa from a 15% hydrogal Number Treatment Extent f%) Adhesion of Animals Meaa, SEMI Qrade,0-2 5 10 A Control 79, 3.6 2,2,2,2,2, , 2,2,2,1,1 7 B Gel 36, 3.7 2,2,1,1,1,1,1 7 C Gel + Ancrod* 10, 2.3 1,1,1,1,1,1,1 *The total dose of ancrod was 50 units/kg 10 total.
~ P < 0.001 by Kruskal-Wallis test.
Use of a photopolymerizable gel with no ancrod reduced adhesions to 25.5 t 7.2%. Uae of-a gel containing 10 units of Ancrod per kg further-15 reduced adhesions to 13.7-t 8.9%.
In saline controls, adhesion was 80.5 t 11.4%
of the extent of the uterine horn. With one IP -injection of 10 units/kg body weight of ancrod on the day of operation, adhesion was reduced to 63.8 t 20 13.4%. Three injections, on days zero, one and two, reduced adhesions to 55.9 t 11.9%. Both effects were statistically significant.
Use of a photopolymerizable gel with no ancrod reduced adhesions to 25.5 t 7.2%. Use of a 25 gel containing 10 units of Ancrod per kg further-reduced adhesions to 13.7 t 8.9%. These reductions were each both statistically significant and clinically relevant.
Exempla 4. Use of Hirudia to Prevent Adhesions.

30 Hirudin is a small protein (7000 daltons) from the medicinal leech -(Merck Index 10th ed #4613).
Its mechanism of action in preventing clotting is to bind tightly to thrombin and prevent thrombin from exerting its enzymatic activities in the clotting cascade, thus ultimately preventing the deposition of fibrin. A refined natural material was obtained from Sigma. A version produced in quantity by standard VJO 95115747 , , PGTlU594/I42t3 fibrin. A refined natural material was obtained from Sigma. A version produced in quantity by standard genetic engineering techniques Hirulog) has essentially identical properties.
Four groups were set up: a control; the gel alone; the gel with 1000 units of hirudin/kg body weight; and, as a further control, the gel with 300 unita/kg heparin.
In experiments conducted essentially as in Example 3, the control had 86% adhesions; the gel alone had 50% adhesions; the gel with 1000 units of hirudin/kg body weight had 19% adhesions; and, as a further control, the gel with 300 unita/kg heparin had 41% ac~hesiona. Thus, hirudin has a significant effect compared to gel alone. Heparin also has a significant effect compared to gel alone, although not a large enough effect to be useful clinically.
Modifications and variations of the present invention, a topical gel delivery system for fibrinolytic agents for prevention of surgical adhesions, will be obvious to those skilled in the art from the foregoing detailed description. Such modifications and variations are intended to come within the scope of the following claims.

Claims (16)

We claim:

1. A composition for preventing tissue adhesions comprising an effective amount of a fibrinolysis enhancing agent in a topically applied biocompatible, biodegradable polymeric matrix, wherein the matrix is formed by gelling or solidifying a biocompatible, biodegradable polymer or macromer on tissue, and wherein the matrix is capable of conforming to the tissue and releasing the agent in a controlled manner over a period of time effective to prevent adhesions.

2. The composition of claim 1 wherein the polymeric matrix is formed by applying the fibrinolysis enhancing agent and a polymerizable.
biocompatible, biodegradable macromer to a tissue site and then crosslinking or polymerizing the macromer, thereby to form the polymeric matrix comprising the fibrinolysis agent in the form of a barrier coating on the tissue.

3. The composition of claim 1 or 2 wherein the fibrinolysis enhancing agent is selected from the group consisting of urokinase, thrombolytic agents, proteases specifically cleaving fibrin, anti-plasmin inhibitors, and fibrin deposition blockers.

4. The composition of claim 1 or 2 wherein the agent is selected from the group consisting of urokinase, hirudin, and ancrod.

5. The composition of claim 1-4 wherein the polymeric matrix releases agent over a period of between a day and thirty days.

6. The composition of claim 1-4 wherein the polymeric matrix degrades over a period of between one and thirty days.

7. The composition of claim 2-6 wherein the matrix binds to the tissue during the polymerization or crosslinking.

8. The composition of claim 1-7 wherein the matrix is formed by free radical polymerization of the macromer.

9. The composition of claim 8 wherein the polymerization of the macromer is initiated by exposure to ultraviolet or visible light.

10. The composition of claim 1-7 wherein the matrix is solidified by a change in temperature or ionic environment.

11. The composition of claim 1-10 wherein the composition is a suspension of polymeric microparticles.

12. The composition of claim 1-11 wherein the matrix is a hydrogel.

13. The use of the composition of claim 1-12 for inhibiting tissue adhesions.

14. The use of the composition of claim 1 for inhibiting tissue adhesions, wherein the composition further comprises a pharmaceutically acceptable carrier for topical administration to tissue.

15. A composition for preventing tissue adhesions comprising a pharmaceutically acceptable carrier for topical administration to tissue and an effective amount of a fibrinolysis enhancing agent to inhibit adhesions, wherein the fibrinolysis enhancing agent is selected from the group consisting of urokinase, thrombolytic agents, proteases specifically cleaving fibrin, anti-plasmin inhibitors, and fibrin deposition blockers.

16. The use of the composition of claim 15 for inhibiting the formation of tissue adhesions.