CA2197244A1

CA2197244A1 - Wound healing formulations containing human plasma fibronectin

Info

Publication number: CA2197244A1
Application number: CA002197244A
Authority: CA
Inventors: Andre Beaulieu
Original assignee: Individual
Current assignee: Individual
Priority date: 1995-06-07
Filing date: 1996-06-06
Publication date: 1996-12-19
Also published as: DE69634430D1; DE69634430T2; EP0784481B1; EP1563844A3; ATE290397T1; US5641483A; ES2236733T3; PT784481E; WO1996040221A1; EP0784481A1; EP1563844A2

Abstract

Topical gel and cream formulations containing human plasma fibronectin are used for the healing of cutaneous wounds. The formulations provide slow release and increased contact time of fibronectin to the wound site leading to effective absoption of an effective wound healing amount of fibronectin in the skin.

Description

096140221 PCT/CA9~00384 WOU~D HEALING FORMU~ATIONS CONTAINING H~MAN PLASMA

.
FIELD OF T~F lN V ~1~ L lUN
The present invention relates to topical dosage forms rrntiq;n;ng human plasma fibronectin for use in promoting wound healing in humans. In particular, the invention relates to the healing of chronic venous ulcers.

R ~ ~ - - OF THE
Fibronectin i9 a large glycoprotein cnnti~;n;ng around 5 carbohydrate. The characteri8tic form of plasma fibronectin is a disulfide-bonded dimer of 440,000 daltons, each i~ubunit having a molecular weight of about 220,000 daltons. Normally found in plasma at a cnnr~ntration of about 300 ~g/Ml, fibronectin is extracted and purified using a method described by Hynesl. Plasma fibronectin is also known by various other names, including cold-insoluble globulin, anti-gelatin factor, cell attiqrl protein, cell spreading factor, and opsonic ~2-surface binding glycoprotein. These names reflect biological activities of _ibronectin such as cell recruitment, op8r,n; 7iqt;nn of particulate debris, and promotion of wound contraction. Reviews on structure and activities of fibronectin have been pnhlllqh~d elsewhere'~3.
~ ound healing is usually divided into three phases: the ;nfli tory phase, the proliferative phase, and the remodeling phase. Fibronectin has been reported to be involved in eac~ stage of the wound healing process, particularly by creating a scaffold to which the invading cells can adhere. Initially, many mediators, such as fibronectin and:fibrinogen, are released to the wound site.
Fibronectin promoteg ;nfli tory cells migration into the wound and debris phagocytosis by the monocytes. Thereafter, angiogenesis and reepith~ ql;7iqticn take place. At this stage fibronectin exerts chemotactic activity on endothelial cells, and promotes the migration of epithelial cells and fibroblasts onto the basal membrane.
~ibronectin also appears to be an essential component of the ? ' 9t244 -- 2 -- ~
remodeling phase where it plays a major role in the organization of collagen fibrils. The fibrillar collagen ultimately forms fibrous bundles that greatly enhance the tissue tensile strength, leading to wound closure.
Topically applied plasma fibronectin has been reported as being useful for increasing the rate of wound healing such as in corneal wounds~S and leg ulcers6. However, no one has described a suitable topical carrier for use in treating wounds that can ensure the delivery of an effective amount of fibronectin. A major limiting factor in developing an effective topical dosage form of a drug is not only having an active drug, but also having a formulation that allows the passage of the active drug from the carrier ~cream, n;ntm~nt, gel, etc.) into the site of delivery (which in the case of the present invention is a skin wound). Very active drugs, such as growth factors, may have no therapeutic value if the topical f~ l~tinn does not allow the drug to move from the semi-soIid carrier into the wound. Therefore, it would be highly desirable to develop a formulation which would maximize the contact time of the fibronectin with the wound and also control the release of f;hrnn~t;n to the wound, thereby leading to high absorption values. The present invention provides such delivery system in the form of aqueous gels and a cream.

8UMMARY OF THE ~w V ~
The present invention provides aqueous gel formulations and one cream fuL i1~t;on ~nnt~;n;ng fibronectin and their use for the delivery of an effective wound healing amount of fibronectin to a wound site. The gel formulation comprises a water soluble, pharmaceutically acceptable polymer which is prepared from an effective amount of f;hrnnP~tin~ Examples of such ~ ~onn~q include: vinyl polymers, e.g. polyacrylic acid; polyoxyethylene-polyu~y~u~ylene block copolymers, e.g.
pnl~ -r; and r~11n1 nse derivatives, e.g.
hyd~u~y~lu~ylcellulose (HPC). The polymer provides viscosity values between 50,000 and l,OOO,OOO cps at room temperature.
The cream fu~ 1~t;nn is prepared from a commercially ?~97244 ~ W096/40221 PCT/CA96/00384 available cream base i.e., Schering~ base (Schering Canada Inc., Point-Claire, Quebec), which has viscosity values between 60,000 to 80,00D cps at room temperature.
Many advantages are attributed to these dosage forms.
Gel and cream formulations of the present invention release effective amounts of a wound healing promoter. Other advantages of gel formulations include: ability to keep the wound moist (which results from the high water content of the gels), ease of application and removal (by washing) from the wound. They also provide a cool feeling when topically applied which can increase patient comfort.
The slow release system of gel formulations of the present invention provides extended release of fibronectin to the wound site. This property of these formulations permits less fre~uent application to the wound resulting in less disturbance to the healing process. Such formulations ~;nt~;n fibronectin delivery for up to 24 hours; but according to kinetic data obtained from p~ ~t;nn studies, a ~twice a day'~ therapeutic schedule is a preferred embodiment o~ the present invention.
Formulation of topical dosage forms intended for the incorporation of fibronectin should respect several ~uality criteria. A11 components of the preparation including solvent, gelling agent and preservative should be nontoxic for the wound and compatible with the drug. The final product should promote optimal release of the drug to its site of action, be of adequate consistency to enhance contact time of the drug with the wound and be sterile.
The preferred formulations of this invention can be used with other wound healing promoters having a composition similar to fibronectin, such as proteins of similar size (thrombospondin, laminin, vitronectin, fibrinogen) or smaller size (such as peptides including growth factors).
The preferred formulations can be correlated with the results of evaluating the formulations using an in vitro ~; f fn~i~n cells system consisting of a rigid receptor c~nt~;n;ng a deepith~ i7e~ skin sample, the deepith~ 1;7ed side facing upwards into a donor compartment , 2 1 9724~
WO96/4022l PCT/CA96/00384 -and the dermal side facing downwards into a receptor compartment. The receptor compartment ia connected to a circulating buffer circuit, with the buffer temperature maintained at 37~C while the skin surface is at about 32~C.

BR}EF DES~K~ . OF T~E FIG~RES
Figure 1 depicts the cumulative absorption of radio-labelled fibronectin over time from various gel formulations rr,ntA;n;ng Carbopol P-934 (0.375~), Pluronic F127 (20.0~), sodium carboxymethylcellulose (CMC 3.0~), and from the control (phosphate buffered~saline solution).
Figure 2 depicts cutaneous absorption of radio-labelled fibronectin from various dosage forms and from the control (phosphate buffered saline solution) at time = I2 hours.
Figure 3 depicts the electrophoresis of human plasma (FN) incorporated in a Carbopol gel (Carbopol P-934 0.375~ +
chlorocresol 0.1~) after 0, 2, 6, and 8 months. Section A:
Recovery of FN after a gelatin-binding test. Section B:
Integrity of FN after 240~days of storage in gel at 4~C. It should be noted that in section B, the resolution of the band is affected by the presence of~r~nt~m;nAnts such as Carbopol in the specimen.
Figure 4 shows a plot of dermal absorption versus viscosity from different topical prepAr~t;~nc.

DTeT~TT.T~n DES~Kl~Ll~_. OF THE lNV
The present invention provides dosage forms that are specially formulated for the therapeutic use of fibronectin as a topical wound healing promoter. ~The dosage forms selected for topical applications should ideally release large amounts of fibronectin, be sterile and non-toxic for the wound. Several factors such as physico-chemical properties of the glycoprotein as well as clinical utilization criteria must be rrnCl ~red when compounding these full ll A t i~n C, Among these limitations, the major one relates to the solubility of fibronectin in water which is poor and therefore mitigates against the preparation of rnnr~ntrated '~1 q7244 ~ WO 96/40221 PCT/CA96/00384 gels or creams. Fibronectin is only slightly soluble in water and may precipitate at concentrations as low as 5 mg/mL
in a~ueous solution. Its solubility is also affected by pH
changes and low temperatures. In the same way, formulations that re~uire the dispersion of polymer powder in the fibronectin solution under agitation cannot be prepared easily since precipitation of the glycoprotein may occur.
~nder agitation, fibronectin may aggregate and form long mats of insoluble material. Viscosity must be optimal in order to permit a sufficient adherence to the wound as well as good release capabilities.
Temperatures over 60~C, which are frerluently rer~uired to provide sterile preparations, denature fibronectin. Since a terminal sterilization process cannot be performed on the final product, the preparation of:concentrated bases of vehicles without fibronectin is usually unavoidable.
Portions of these sterile bases are then diluted with a definite amount of a solution of fibronectin. To achieve adequate dispersion of fibronectin into seml-solid dosage forms, an incorporation step involving agitation i9 often re~uired which san lead to the precipitation of the drug.
Gelling agents such as ~rhopol and pol~ can ciliu~l~v~llL this problem since they are sterilized before gelation under a li~uid-like, viscous form. A highly r~nr~ntrated preparation of Carbopol is prepared and autoclaved. A8 described below, the solution of fibronectin which also c~nt~;nR the polymerization promoter (NaOH) is then mixed in syringes with the base of Carbopol, bn;l~;ng-up the gel during the dispersion of the drug into it. In the case of pol~ -r, the polymer is added to the drug solution and allowed to dissolve at 4~C, a temperature at which it ~;nt~inc its fluid-like aspect. Sterilization of this solution from bacteria is performed at 4~C using a 0.22 ~m ~ filter.
A non-toxic, non-sensitizing preservative ~-t;hl~
with for~ulation , _ ~nt~ is added to the dosage form in a preferred ~ ~;r ' of the invention. All of the above c~n~;t;~nR are respected in the preferred dosage forms described in detail a~ follows.
An effective wound healing amount of human plasma fibronectin for use in the present invention is within the range of 0.05 to 0.~ by weight and more preferably between 0.2 and 0.4~. Fibror,ectin i5 isolated from human plasma by using a gelatin-Sepharose affinity ch~ ~ tsgraphy procedure.
In this method, gelatin is covalently coupled to Sepharose 4B
after CNBr activation. The binding capacity for human plasma fibronectin provided by this system is ~ 1 mg/ml of gel.
Autologous, homologous human plasma fibronectin or fibronectin obtained from recombinant-DNA technology may be used in the present inventionl~7. Shoul~ homologous plasma f;hrnn~tin be used, lots prepared from different donors would have to be tested for atypical antibodies, hepatitis B
~BV), hepatitis C (~CV), human immunodeficiency virus (HIV), human T-cell lymphotrophic virus (~T~V), cytomegalovirus (CMV) and syphilis. Thefie tests have to be performed on donors just prior to the ~nn~t;nn and 6 months thereafter.
In the meantime, donor plasma mu3t be kept frozen at -20~C.
Furthermore, special steps should be taken in order to inactivate potential viruses. An inactivation method using tri (n-butyl) phnRrh~tP/Tween-80 or tri~(n-butyl) phosphate/Triton X-100 (solvent/detergent method) should be performed on all plagma ~nn~t;nnR7~.
In the gel formulation for topical wound healing, the viscosity may be within the range of 50,000 to 1,000,000 cps, more preferably between 100,000 and 650,000 cps. In the cream formulation, the viscosity may be within the range of 60,000 to 80,000 cps. All viscosity values are in centipoises (cps) as measured using a Brookfield viscometer.
Assays were performed at 0.5 rpm and at room temperature.
In one embodiment of the present invention, the gel formulation may comprise 0.25 to 1.0~ by weight polyacrylic acid having a molecular weight of about 740,000 to 5,000,000.
In a preferred '~';~~ ~, the polyacrylic acid is present at 0.35 to 0.75~ by weight and has a viscosity of about 350,000 cps. The p~ of the polyacrylic acid gel should be within the range of 5 to 8 and more preferably between 6.5 and 7.5.

~t ~7~
~ WO96/4022l PCT/CA96/00384 Polyacrylic acid iB commonly known as Carbopol and the preferred grade iB P-934.
In another embodiment, the gel formulation may comprise 18 to 35~ by weight polyoxyethylene-polyu~y~.u~ylene block copolymer having a molecular weight of about 2,000 to 13,000.
In a preferred embodiment, the polyoxyethylene--polyu~y~lupylene block copolymer is present at 18 to 25~ by weight and has a viscosity of about 450,000 cps at room temperature. The pH of the block copolymer gel should be within the range of 6 to 8 and more preferably between 6.5 and 7.5. Polyoxyethylene-polyoxypropylene block copolymers are commonly known as Pluronic and the preferred grade is E-127 (poloxamer 407).
In a further '-'; , the gel formulation may comprise 1 to 5~= r~ nRe derivative which may be hyd,u~y~L~ylcellulose (HPC) and has a viscosity of about 25,000 to 150,000 cps. HPC has a molecular weight of about 370,000 to 1,150,000. In a preferred Pmho~; t, the rrl 1 1l1 o-Re derivative ig present at 2.0 to 4.0~ by weight and has a viscosity of about 150,000 cps for HPC. Cellulose derivatives used in the present invention are commonly known as Klucel for HPC. Preferred grade is Klucel-HP.
In a further embodiment, a cream formulation is prepared from a commercially available cream base i.e., Schering~
base. This cream base (oil in water ~ lR;nn) ~nn~R;nc ceteth-20, cetostearyl alcohol, chlorocresol, mineral oil, monobasic sodium phosphate, phosphoric acid, sodium hydroxide, water and white petrolatum. The viscosity of the preparation can be modified by varying the content of water and polyethylene glycol.
Formulations of the present invention contain an aqueous phase in combination with a protein and thus are prone to attack by bacteria and fungi. Microbial growth not only spoils the=formulation but is a potential toxicity hazard and a source of infection for patients. Even though microbial growth is less likely to be dangerous when it occurs in a topical preparation, it is especially important to preserve topicals which patients have to apply to broken or inflamed ~ . .

2 1 ~724~
WO96/4022l PCT/CA96/00384 -_ ~ _ skin. Viscosity degradatlon reported with~some polymera when exposed to microbial contamination is also of concern~ So, a preservative should be added to the preparation to guarantee long term sterility and stability. The present invention provides gels that comprise a preservative selected from phenol or the para-hydroxybenzoate compounds. In one embodiment, the gel formnlAt;~n may contain 0.1 to 0.29D by weight chlorocresol, a phenol derivative or 0 01 to 0_3~ by weight p-hydroxybenzoate as methyl- and propylparaben= In another embodiment, the cream formulation contains O.1 to 0.2~ by weight c_lorocresol.
St~h;li7~rs may be added to the formulation in order to provide stable compositions of fibronectin. They may help to preserve biological activities on a long term basis and may improve water solubility of fibr~n~rt;n Among these agents, albumin, ~;qacrhArides such as sucrose, and cyclic olig~q~rrhArides such as cyrlo~p~tr;nq are stabilizers of choice These agents can be used either alone or in _ combination. Human albumin is preferable in terms of antigenicity and should be free from microbial r~ntAm;n~t;~n Cyrl~P~trins of the ~ group (7 glucose units) are of choice and hydroxypropyl-~-cyclodextrin is pr~f~rAhl~. The formulation may comprise o.01 to 0 196 by weight albumin, preferably 0.01 to 0.05%; and/or 0.5 to 5_09O by weight sucrose, preferably 3.0 to 5 09~; and/or 1.0 to 10% by weight hydroxypropyl-~-cyclodextrin, preferably 2.0 to 5.0~.
Some authors have suggested that protease activity in some chronic wounds may cauRe degradation of adhesion proteins such as fibronectin and prevent cell A~h~q;~n r~r~qq~ry for normal wound closure9. Metalloproteases and serine proteases have been ;~nt;fied in chronic wound fluid9~l~ and fibronectin has been reported to be highly sensitive to cleavage by proteases1l. ~rotection of the integrity of fibronectin may be accomplished by the addition of protease ;nh;h;t~rs in the dosage form. The present invention also provides formulations that may comprise a r t~lloprotease inhibitor such as EDTA and/or a serine protease inhibitor such as aprotinin (Trasylol~, Miles) with -2 ! 97244 ~ W096/402~1 PCT/CA96/00384 _ g this aim in view. ln one embodiment, the dosage form may comprise 0.01 to 1.0~ by weight EDTA and/or 1.5 to 45.0 Inh U
~ by weight aprotinin where 1 Inh U = 26 Kallikrein inhibitor units.
Formulations of the present invention can be applied to the wound~site by any suitable means which assures that the wound surface will be entirely covered. For example, it can be directly applied to the wound site or used to coat fibers of an absorbent gauze dressing to form a wound healing bandage which may then be placed on a wound.
Examples which follow are ;nt~n~rd to illustrate furt~er aspects of the invention and are not to be construed as limiting its scope in any way.

~ P~E 1 Isolation of fibronectin from 1 ~laama 1) A ater;l;7~tion step is mandatory for all homologous plasma d~nAt;onp. In order to inactivate potential viruses, a eterilization procedure using the 8 solvent/detergent method is ueed. 1~ tri ~n-butyl) phosphate (TNBP) and 1~ Triton X-100 are added to the plasma for 6 hours at 24~C. After that, soybean oil i9 added to the plasma and allowed to be mixed for at least 30 minutes in order to extract TNBP.
R~R; ~ 1 Triton will be eliminated by dialy8is. I~i~
f;~st 8tep is 8kip~ed if autolo~ou8 ~1~P~- is used.
2) A gelatin-Sepharose 4B column is first p~ rl with a Tria-HCl solution in order to equilibrate the gel.

3) The plasma is diluted ~1:1) with a Tris-HCl solution and pumped through the column in the presence of phenylmethylsulfonyl fluoride 0.001 M ~PMSF) for about 15 hours at 4~C.

4) The column is then washed three times in order to elute nonspecifically bound plasma proteins from the gel. All washing steps are performed using a Tris-HCl pH 7.5 solution. A 1 M NaCl solution is added to the second washing step to elute c~nt~m;n~ntP.

5) Elution of f;hrrn~r~tin is carried out by using 0.1 M

~ =

2~1 q72~4 W096/4022l PCT/CA96/00384 -Na acetate + 1 M KBr solution.

6) Two dialysis steps are then performed to eliminate contaminants ~Triton X-100, RBr, Na acetate) ~ialysis ver~us PB5 and sterile water are respectively done.

7) Solution is concentrated by ultrafiltration under nitrogen pressure.

8) Terminal sterilizing filtration using a 0.22 ~m filter is done to warrant sterility.

9) Fractions are aliquoted and frozen at~~-20~C until their incorporation into the topical dosage form.

~MPLE 2 PolYacrylic acid ~els Polyacrylic acid (~rhl_ ~r) gels (rArhornll~ BF
Goodrich) were prepared. r~ -- (carbopol is a polymer derived from acrylic acid. It is a high molecular weight polymer (740,00 to 5,000,000) that gelifies when neutralized by strong alkalis (NaQH) or amines ~triet~n~l~m;nP). It forms gels at relatively low c~nc~ntrations, that is as low as 0.25~, and its viscosity is strongly reduced by the addition of electrolytes.
Preferred grade of polyacrylic acid is Carbopol 934-P at c~n~ntrations ranging from 0.3~ to 0.75~ ~w/w). Lower r~n~ntrations are insufficient to promote~adherence to the wound and higher c~n~ntrations reduce the release of fibronectin from the gel. Viscosity of polyacrylic acid gels is stable between p~ 6 to 8 with a preferred pH range between 6.5 to 7.5. Viscosity is reduced in the presence of strong electrolytes.
A polyacrylic acid gel ~nt~in;ng (w/w) fibronectin 0.2~, r~rhopol 934-P 0.375~, and chlorocresol 0.1% was prepared as follows: chlorocresol (1.0 g) was dissolved in~
warm (65~C) ~ n;7~ water (95 mL) under slow agitation.
When the chlorocresol is completely dissolved, the solution is cooled at room temperature while ~-;nt~;n;ng agitation.
Carbopol 934-P (3.75 g) was then added, dispersing it slowly on the surface of the s~ t;~n, and mixed with a paddle type ~ ~ 97244 g6/4022l PCT/CA96/00384 stirrer for about 3 hours. This dispersion was then autoclaved to provide a sterile concentrated gel base (3.75~
w/w). A stock solution of fibronectin 2.2 mg/mL (90 mL) was filtered t_rough a 0.22 ~m acetate filter. A polymerization promoter, sodium hydroxide, was added to the fibronectin solution in an amount that will neutralize a 10 g portion of the Carbopol 3.75~ dispersion, that is 1250 ~L of NaOH 3M.
The stock solution of fibronectin and Carbopol dispersion were mixed into syringes taking care-to avoid introducing air bubbles and cnn~Am;n~icn in an aseptic environment, such as under a laminar flow hood. This preparation provides a clear, preservea gel (100 g) of fibronectin free from microorganisms with viscosity of about 350,000 cps.
This gel fo l~tion was appliçd twice a day on leg ulcers in a pilot study in humans and showed an Pnh~nc~d rate of wound healing without any adverse effect.

R~PLE 3 Polvoxvethvlene-Pol~u~y~ruuvlene block co~olvmer qels Polyoxyethylene-polyoxypropylene block copolymer (poloxamer) gels (Pluronic~, BASF Wyandotte) were prepared.
Preferred grade of poloxamer is Pluronic F-127 at concentrations ranging from 18 to 25~ (w/w). Poloxamer F-127 is a low molecular weight polymer (2,000 to 13,000) which exhibits thermal gelation characteristics. Gelation occurs when the concentration reaches 18~ poloxamer. The viscosity of po1~ r ic proportional to the rnnrPn~raticn of the polymer, type of polymer used (molecular weight) and temperature. Fluid at 4~C, the polymer gelifies with increasing temperatures, providing high viscosity values at room temperature. In contrast to Carbopol, the addition of ions ~nh~nrPs the viscosity of the preparation.
~ nnr~n~rated aqueous solution (20 to 30~) have been reported to show a dramatic increase in viscosity when heated from 4~C to body temperature. Furthermore, if the ionic strength of the solution is increased, the viscosity is increased more rapidly with rising temperature. Several grades are available but the F-127 grade is the least toxic 2 ~ ~7244 WO96/4022l PCT/CA96/00384 and gelation can occur at lower concentrations. Gels of poloxamer prepared in this invention are low viscosity solutions at 4~C and gelify rapidly when they are warmed to body temperature.
A poloxamer gel containing (w/w) fibronectin 0.2~ and Pluronic F-127 20~ was prepared as follows: a stock solution of fibronectin 2.2 mg/mL (80 m~) was filtered through a 0.22 ~m acetate filter. Pluronic F-127 (20 g) was added to 80 mL
of the fibronectin solution and allowed to dissolve without agitation at 4~C ior about 3 days. The resulting solution (100 g) is very li~uid-like. Gelation occurs instantly when the solution comes into contact with the wound. A
sterilizing filtration process performed at 4~C could~also be applied to the final solution if sterile poloxamer powder cannot be obtained. ~iscosity varies from not detectable values at 4~C to 450,000 cps at room temperature.

EXAMP~ 4 Cellulose derivative ~els HydLu~y~Lu~ylcellulose (XPC) gels were prepared~ n order to illustrate this type of formulations, the preparation of a HPC 3~ gel is described as below. Preferred grade is ~lucel-HF at cnn~ntr~tinnq ranging from 2 to 4~
(w/w) .
A gel formulation nnntA;n;ng (w/w) fibronectin 0.1~, HPC
3~ and parabens was prepared as follows: methylparaben 10.05 g) and propylparaben (0.02 g) were dissolved in warm ~;nn;7e~ water l94 m~). HPC powder was sterilized by using a dry heat st~r;l;7~t;nn process. HPC (6 g) was then dispersed in this solution and allowed to be mixed with a paddle type stirrer for about 3 hours. This provides a sterile concentrated gel base (6~ w/w). A stock solution of fibronectin 2 mg/m~ (50 m~) was filtered through a 0.22 ~m acetate filter. Fibronectin solution (50 mL) was then slowly added to=a portion (50 g) of this concentrated base using the low-speed shaft of the stirrer. This provides a preserved gel (100 g) with viscosity of about 150,000 cps.

21 97~244 ,R~Al~qpLE 5 t~rf~m forrlll Ation A cream formulation ~nntA;n;ng (w~w] fibronectin 0.1~, sterile cream base (Schering~ base, Schering) and 5 chlorocresol 0.1~ was prepared aa followe a stock sol~ltinn of fibronectin 2 mg/mL (50 mL) was filtered through a 0.22 ~m acetate filter. Fibronectin solution (50 mL) was then added filowly to a portion (50 g) of the cream base using the low-speed shaft of a stirrer. Thie p~ovides a preserved cream (100 g) with viscosity of about 70, 000 CpS.

Rl~l~MPT.R. 6 Ki~etics of release from ~;fferent to~ical dosa~e forms The effectiveness of each topical f~L lAtin~ to release fibronectin was evaluated using an in vitro diffusion cell system. Permeation studies were all performed on human breast and Ah~ nAl deepith~1;A1;~ skin samples nhtA;n~d from breast. reduction and Ah~t nAl l;rectl y surgeries. A 8 ~m section was removed from the epidermal surface of the skin using a dermatome (1/10,000 scissor scale) and the dermal side was carefully cleaned of any a &ering enhClltAn~nn~
tissues and/or blood vessels. DeepithP1;A1;7~d human skin was used in order to reproduce the pathological condition met in chronic venous ulcers where the epidermis layer is absent.
The diffusion cell system selected consisted of a rigid 25 receptor contA;n;ng the skin sample, the deepi~h~1;A1;~ed side facing upwards into the donor c _--i and the dermal side facing downwards into the receptor compartment. The receptor compartment was cnnn~ct~ to a circu1A~;rg buffer circuit. The buffer temperature was r-;ntA;n~ at 37~C while 30 the skin surface was at about 32~C. Each analysis was performed on a 0. 64 cm2 skin sample using a 100 ~L aliquot of "5I-fibronectin topical formulation spe~; . After the experiment, the skin was removed from the diffusion cell, washed 10 times with a 8 mL water volume by wash, and 35 analyzed for its content of radioactivity in a gamma radioactivity counter. The total amount Ahsorh~d (dermis +
receptor compartment) divided by the dose applied gave the wo 96/40221 PCT/CA96/00384 percent ab~orption.
All dosage forms were made in salt-free solution since viscosity values could have been influenced by the presence of electrolytes. For instance, viacosity values of carbomer 5 gels are reduced in the presence of strong electrolytes in contrast with poloxamer gels which are more viscous when electrolytes are added to the preparation.
Several authors have compared percutaneous absorption studies using in vitro and in vivo techniques to establish 10 the rPl;~hil;ty of results using these method3l3 1'l5. These comparisons have clearly shown that in vitro studies can accurately reflect the living state. 3tatistic analysis applied to our experiments has demonstrated a good correlation value between studies performed on skin obtained 15 from different sources. These data have shown that the origin of the skin did not have any effect on results.
Percutaneous absorption studies are usually performed on intact skin and are designed to evaluate the release of a substance from a topical vehicle and its absorption through 20 the major cutaneous barrier, that is the stratum corneum. In cutaneous ulcers, the barrier effect of the stratum corneum is absent. With this pathological condition, only the diffusion from the dermatological vehicle will be a major detPrm;n~nt for the ulterior penetratio~ of the drug into the 25 dermis.
The difiusion cell system described above i~ a suitable in vitro model for ,-n~nPml~ ulcers.
Rinetic data of the release of fibronectin from various dosage forms were nht~;nP~ at 4, 12, and 24 hours. Table 1 30 summarizes the~3e data for t=12 hourc. The control consisted of l2sI-fibronectin in phosphate buffered saline solution, pH
7 4 T~;ros -~ uced in the ~'~rhopol 934 P (1%) + liposomes (15$) formulation (Lipogel) were made from Proliposomes (Pro-lipo 3090 SHT'I, Lucas Mayer, France). Cellulose derivatives 35 are identified as CMC for sodium carboxymethyl~ lnse and HPC for hydLu~y~LlJ~ylcellulose. Dermabase~ (Bcrdenl Ltee., Don Mills, Ontario, Canada) and Scheringl base are cream bases available on the market and were diluted 1:1 for these 2 1 972~4 ~ W096/40221 PCT/CA96/00384 experiments. The symboY [ ] refers to componentsconcentration and "Abs value" to the percentage of radiolabelled fibronectin found in the dermis after an exposition time of 12 hours.
TA3L~ 1 F~ Ab~ v~l~
Control 24.75 Lipogel 3.70 Dermabase tl:1) 5.80 C~C 3S 6.70 Carbopol 934 P+ glycerol 0.375%/10~ 7.80 (Carbogly) Schering baDe (l:1) 9.90%
l~ Pluronic F-127 20% 12.80 Car~opol 934 P0.375~ 13.40 HPC 3~ 15.20 Figure 1 plots kinetic data of three gel dosage forms and control solution over time. From this graph it can be seen that the absorption process tend to be more important between time 0 and 12 hours than between time 12 and 24 hours, suggesting that two applications per day could release more fibronectin than an once a day schedule.
Figure 2 depicts cutaneous absorption of rA~;olAhelled fibronectin from various dosage forms and from control at time = 12 hours. The Dunnett statistic test was used to identify statistically significant differences between mArhQpo~ and other f~., lAt;~nq. This test has also shown significant ~ r~n~ between Lipogel, Carbogly, and Carbopol, results that can be correlated with those of effectiveness obtained during rl;n;~?l trialg (see example 8). The efficacy of the - Carbopol f-1 1 A~ n i5 particularly surprising since Carbopol has a higher degree of viscosity than many of the other formulations studied. Also noteworthy are the difference in Abs value between the Carbopol and CMC
formulations since they both share the same degree of viscosity.

~1 97244 Flgure 4 show that a clear relat;nn~h~p between viscosity and absorption does not always existL when , considering the some of the preparations for which viscosity values were determined. For~instance, Dermabase whic~ has a relatively low viscosity (119,000 cps) when compared to Carbopol (411,300 cps) presents poor release r~p~hilities (5 30~) when compared to Carbopol (13.40~).

E~PLE 7 StAh;1itv of fibronectin in qel Biological activity and integrity of the macrostructure of fibronectin in gel formulations were evaluated (Figure 3). Assays were performed on a specimen of gel c~nt~;n;nr~ (w/w) fibronectin 0.2~, Carbopol P-934 0.375~, and chlorocresol 0.1~. The specimen had been kept at 4~C for 32 weeks.
Electrophoresis technigues were used in order to determine the integrity of macrostructure of fihronectin in gel. After the sp~r; of gel was dissolved in lM NaCl +
Tris-~Cl p~ 7 4 solution, it was allowed to migrate on an 7.5~ acrylamide gel according to the method of Laemmli ("Denaturing (SDS) disrnnt;nnnus gel electrophoresis: Laemmli gel method," pages: 10.2.4 - 10.2.9, Current Protocols in Molecular BioloqY 1994). Compared with a fresh standard solution (column 0), results showed that close to 100~ of the fibronectin can be i~Prt;f;ed around the 220,000 (column B) band indicating that very little, if any, degradation occurs.
R;~1og;r~1 activity was evaluated using an affinity chromatography test Gelatin-binding is one of these biological activities that can be assessed with relative ease. After a specimen of gel was dissolved in a lM NaCl solution, a known amount of this viscous solution was placed in an Eppendorf tube in the presence of gelatin Sepharose 4B
and then vortexed. ~he conte~t was further rinsed with a fresh lM NaCl solution, centrifuged and the snrPrn~t~nt discarded in order to ~1 i~in~t~ c~nt~;n~nt8 such as Carbopol and chlorocresol that came from the dissolution of the gel.
Fibronectin was eluted from the gelatin-sepharose 4B using a ~97~;~4 1 M K3r solution. The fractlon collected was allowed to migrate on an 7.5~ acrylamide gel according to the method of Laemmli. The band was then evaluated with respect to its content of fibronectin using a densitometric scanning assay.
The specimen collected could also be evaluated spectrophotometrically using optical density at wave length 280 nm.
Compared with~a freshly prepared gel of fibronectin (column 0), it can be seen that a large amount (80~) of fibronectin was recovered from the specimen of gel formulation (column 8 months) indicating that gelatin-binding activity of the glycoprotein can be preserved for a long period of time in this dosage form.

T'U2~1~qpT.li. ~
Cl;n;~Al tr;AlR: treatment of chronic leq ulcers We have conducted four clinical trials (pilot studies) to investigate the usefulness of different dosage forms containing exogenous human plasma fibronectin in the treatment of chronic venous ulcers of lower limbs. In these trials autologous plasma fibronectin was used and patients with ulcers that were resistant to the conv~nt;~nAl therapy for at least three months were selected.
The ~p~r;~;~ objective of the first experiment was to determine the effectiveness of topically applied fibronectin as a wound healing promoter. Seven patients were included in this study and were instructed to "flood~ the wound area with a solution of f;hr~n~tin 1 mg/m~ (0.1~) in PPS (phosphate buffered saline~ twice a day. After two months of regular application of that solution, five of these patients presented with a dramatic decrease in their wound size, specifically at least 75~ r~ rt1nn of the integrated surface area.
A second experiment was designed to evaluate the effectiveness of a semi-solid dosage form which c~ntA;nP~ by weight fibronectin 0.1~, ~ncAps~ ted in liposomes 15~, which in turn were incorporated in Carbopol (1~) formulation known as Lipogel. Hypothesis was that if the contact time of the 2 ~ q7~44 W096/40~t PCT/C~96/00384 -glycoprotein with the wound could be ~n~nr~ a more rapid decrease in the healing time could theoretically be observed.
Six p~ti~nts were included in this study and they had to apply the formulation to their wound twice a day. None presented a substantial decrease of their wound size during the following three months of regular trea~ment.
In an attempt to improve the dosage form, an experiment was undertaken to evaluate the therapeutic potential of a topical gel formulation rnnt~;n;ng (w/w~
fibronectin 0.2~ incorporated in Carbopol 0.375~ and glycerol 10~ ~Carbogly). Glycerol had been added to the formulation in order to take advantage of its humectant effect which could be beneficial to the wound. Eleven pat; ~nt q were recruited for this study and they also had to apply the gel twice a day. Among these patients, 27~ had a regression of more than 50~ of their wound size after three months of treatment.
Results from the permeation studies may explain, at least in part, what could have occurred in previous exp~r;r ~. Figure 2 shows that preparations such as ~ipogel and Carbopol + glycerol do not lead to high absorption values. In contrast, Carbopol 0.375~ without glycerol provides significantly higher absorption values (p<0.001~. ~he solution used in the first ex eriment is ;~rnt; f ied as the control in this graph. This last preparation provides the highest release r~r~h;l;ties but it does not represent a forr~ tinn that could be useful to patients owing to its fluid consistency.
Con~idering these results, a formulation rnnt~;n;ng fibronectin 0.2~ (w/w) in Carbopol 0.375~, without glycerol was investigated in eight patients. According to clinical and permeation studies, this formulation is the preferred carrier using Carbopol that is available for the use of f;hrnn~rtin in topical wound healing. Pr~l;m;n~ry data showed that 5D~ of patients studied presented a regression of more than 5Q~ of their wound size within three months of treatment, including two complete responses (100~ healing) that occurred within the first eight weeks of treatment. The ~-2 ~ 972~4 ~ WO96/40221 PCT/CA96/00384 present invention also provides other formulations that are as usefuI as thIs one using the permeation study described in Example 5 as a model system to test the various formulations.

El~hMpT.R 9 ~re re~orts To illustrate the efficacy of the formulation rnnt~;n;ng fibronectin and Carbopol 934-P 0.375~ (w/w), we present two s~ecific cases of c,hronic venous leg ulcer.
These cases are of interest in that the first case was highly res1stant to convPnt;nn~l therapy and the second case was a large ulcer.~ Eactors such as~ duration and surface area have been ;~Pnt;fied by several authors as playing a major role in the prognosis of the venous ulcer.

CaDe 1 A 37-year-old woman presented with a ten-year history of chronic venous ulcer of the right lower limb. Eer medical history was not significant except for four episodes of phlebitis. The last episode occurred during pregnancy and ultimately resulted in an ulcer. Review of medical trP~ ~ that were tried revealed the use of topical antiseptics, elastic stockings, and skin grafting without any positive result.
The patient presented to our clinic with a 1.60 cm2 aching wound ~espite the fact that her ulcer was relatively small, it appeared highly resistant to therapy. Six weeks after starting the application of the gel of fibronectin, a 92~ r~nrt;nn o_ her wound size could be observed. Complete reepit~Pl;~l;7At;nn was noted after a ten-week course of tLI~ t A follow-up visit scheduled one month later revealed no ~ptpr;nration in her wound rnn~;t;nn.
Ca~e 2 A 39-year-old man presented with a seven-month history of chronic venous ulcer of the left leg. His medical history was not significant except for a ~p~n~ct~ y of the left lower limb twelve years before. Topical antibiotics were prescribed to the patient without any effect on his wound 2 ~ ~7244 W096/4022~ PCT/CA96tO0384 -size.
He presented to our clinic with a 10.5 cm~ ulcer resulting from a local trauma. L~ ph~ of the left lower limb was important and a large crusty necrotic layer bordered 5 the wound. The patient's ~c~nration constrained him to remain standing for long periods of time. Although thi3 situation probably worsened his wound condition, it could not be eliminated.
After four weeks of regular application=of a placebo gel and normal saline, the wound size increased to 21.5 cm2 as a consequence of local debridement. The placebo gel comprised 0.375~ Carbopol 934-P, 0.1~ chlorocresol, purified water and NaOH to adjust the pH. Active treatment with the fibronectin-~nntA;n;ng ~Arh~p~l gel formulation was begun at 15 this time. Maximum wound size was noted six weeks later (37.5 cm'), revealing a larger ulcer than initially assumed.
The wound healing process took place between six to eight weeks and was completed after 31 weeks of active treatment.
A follow-up visit scheduled one month later revealed no deterioration in his wound condition.
Although the present invention has been described in relation to particular : 's~; q thereof, many other variations and modifications and other uses will become apparent to those skilled in the art. It is therefor~
25 understood that numerous variations of the invention can be made which are well within the scope and spirit of this invention as described in the ~rp~n~ claims.

~5~
1 - Hynes, R. O., Methods for identification of fibronectin 30 (chap. 2, page 12), IN: Fibronectins New-York: Springer-Verlag, 1990.

2 - Hynes, R.O., Methods for identification of fibronectin (chap. 2, pages 7-23) and Wound healing, inflammation, and~
fibrosis (chap. 14, pages 349-64), IN: Fibronectins New-35 ~ork: Springer-Verlag, 1990. I

21,97?44 ~ W096/4022l PCT/CA96/003 3 - Brotchie, H., Wakefield, D. Fibronectin: Structure, function and significance in wound healing. Au~tralae J
Dermatol l990; 31:47-56.

4 - Nishida, T., Nakagawa, S., Awata, T. et al., Rapid preparation of purified Ant~logollR fibronectin eyedrops from patient's plasma. Jpn J O~hth~ -l 1982;26:416-24.

5 - Ehan, T.M., Foster, C.S., Boruchoff, S.A. et al., Topical fibronectin in the treatment of persistent corneal epithelial defects and trophic ulcers. Am J Orhth~ l 1987;104: 494-501. : _ 6 - Wysocki, A., Baxter, C.R., Bergstresser, P.R. et al., Topical fibronectin therapy for treatment of a patient with chronic stasis ulcers. Arch Dermatol 1988; 124:175-7.

7 - Edwards, C.A., Piet, M.P.J., Chin, S. et al., Tri (n-butyl) phosphate/detergent treatment of licensed therapeutic 7nd experimental blood derivatives. Vox Sang 1987; 52:53-9.

8 - Horowitz, B., Bonomo, R., Prince, A.M. et al.
Solvent/detergent-treated plasma: a virus-inactivated substitute for fresh frozen plasma. Blood 1992; 79: 826-31.

9 - Grinnell, F., Ho, C.H., Wysocki, A., Degradation of fibronectin and vitronectin in chronic wound fluid: Analysis by cell blotting, ; -l~lotting, and cell ~h~ n assays.
J Invest Dermatol 1992; 98: 410-6.

10 - Chen, W. Y. J., Rogers, A.A., ~ydon, M.J., Characterization of biologic properties of wound fluid collected during early stages of wound healing. J Invest Dermatol 1992; 99: 559-64.

11 - Berman, M., Manseau, E., Baw, M. et al., Ulceration is correlated with degradation of fibrin and fibronectin at the corneal surface. Invest Crhth~l l Vis Sci 1983; 24: 1358-. ~

2' ~72~
W096/4022l PCT/CA96/00384 66.

12 - ~orowitz, B., Chang, M.D.Y., Preparation of fibronectin for therapeutic administration. IN: D.F. Mosher (ed.), Fibronectin, pages 441-55, San Diego, Academic Press 1989.

13 - Franz, T.J., Percut~n~ouq absorption. On the relevance of in vitro data. J Invest Dermatol 1975: 64:1gO-95.

14 - Bronaugh, R.L., Stewart, R.F., Congdon, F.R. et al.
Methods for in vitro percutaneous absorption studies. I:
Comparison with in vivo results. Toxicol Appl Pharmacol 1982;62:474-80.

15 - Bronaugh, R.L., Stewart, R.F., Methods for in vitro percu~nPol~ absorption studies IV: The flow-through diffusion cell. J Pharm Sci 1985: 74:64-67.

Claims

What is claimed is:

1. An aqueous gel formulation for healing chronic wounds, comprising:
a) an effective wound healing amount of human plasma fibronectin;
b) a water soluble, pharmaceutically acceptable polymer having viscosity of about 50,000 to about 1,000,000 cps at room temperature; and c) an effective amount of a preservative.

2. The aqueous gel formulation of claim 1, wherein the human plasma fibronectin is obtained from autologous, homologous human blood or recombinant-DNA technology.

3. The aqueous gel formulation of claim 1, wherein the human plasma fibronectin concentration is within the range of about 0.05 to about 0.5% by weight, based on the weight of the entire composition.

4. The aqueous gel formulation of claim 1, wherein the polymer is selected from the group consisting of a vinyl polymer, a polyoxyethylene-polyoxypropylene block copolymer and a cellulose derivative.

5. The aqueous gel formulation of claim 4, wherein the vinyl polymer is a polyacrylic acid.

6. The aqueous gel formulation of claim 5, wherein the polyacrylic acid has a molecular weight of about 740,000 to 5,000,000 and is present in an amount in the range from about 0.25-1.0% by weight, based on the weight of the entire composition.

7. The aqueous gel formulation of claim 4, wherein the polyoxyethylene-polyoxypropylene block copolymer has a molecular weight of about 2,000 to 13,000 and is present in an amount in the range from about 18 to about 35% by weight, based on the weight of the entire composition.

8. The aqueous gel formulation of claim 4, wherein the cellulose derivative is hydroxypropylcellulose (HPC).

9. The aqueous gel formulation of claim 8, wherein the hydropropylcellulose has a molecular weight of about 370,000 to 1,150,000 and is present in an amount in the range from about 1 to about 5% by weight, based on the weight of the entire composition.

10. The aqueous gel formulation of claim 1, wherein the preservative is selected from the group consisting of chlorocresol and a combination of p-hydroxybenzoate (parabens).

11. The aqueous gel formulation of claim 10, wherein the chlorocresol is present in amount in the range from about 0.1 to about 0.2% by weight and the p-hydroxybenzoate is present in an amount in the range from about 0.01 to about 0.3 % by weight, based on the weight of the entire composition.

12. The aqueous gel formulation of claim 4, which may comprise at least one additional ingredient selected from the group consisting of stabilizers and protease inhibitors.

13. The aqueous gel formulation of claim 12, wherein the stabilizer is albumin, a disaccharide, or a cyclic oligosaccharide, or a combination thereof.

14. The aqueous gel formulation of claim 13, wherein the albumin is present in an amount in the range from about 0.01 to 0.1% by weight, the disaccharide is sucrose which is present in an amount in the range from about 0.5 to about 5.0% by weight, the cyclic oligosaccharide is hydroxypropyl-.beta.-cyclodextrin which is present in an amount in the range from about 1.0 to about 10.0% by weight, all percentages by weight being based on the weight of the entire composition.

15. The aqueous gel formulation of claim 12, wherein the protease inhibitor is a serine protease inhibitor or a metalloprotease inhibitor.

16. The aqueous gel formulation of claim 15, wherein the serine protease inhibitor is 1.5 to 45 inh U % by weight aprotinin or the metalloprotease inhibitor is 0.01 to 1% by weight EDTA, or a combination thereof.

17. A cream formulation for healing chronic wounds, comprising:
a) an effective wound healing amount of human plasma fibronectin;
b) a cream base having a viscosity of about 60,000 to about 80,000 cps at room temperature; and c) an effective amount of a preservative.

18. The cream formulation of claim 17, wherein the cream base comprises ceteth-20, cetostearyl alcohol, chlorocresol, mineral oil, monobasic sodium phosphate, phosphoric acid, sodium hydroxide, water, and white petrolatum.

19. The cream formulation of claim 17, wherein the human plasma fibronectin is obtained from autologous, homologous human blood or recombinant-DNA technology.

20. The cream formulation of claim 17, wherein the human plasma fibronectin concentration is within the range of 0.05 to 0.5% by weight.

21. The cream formulation of claim 17, wherein the preservative is chlorocresol which is present in an amount in the range from about 0.1 to about 0.2% by weight based on the weight of the entire composition.

22. The cream formulation of claim 17, which may comprise at least one additional ingredient selected from the group consisting of stabilizers and protease inhibitors.

23. The cream formulation of claim 22, wherein the stabilizer is albumin, a disaccharide, or a cyclic oligosaccharide, or a combination thereof.

24. The cream formulation of claim 23, wherein the albumin is present in an amount in the range from about 0.01 to 0.1% by weight, the disaccharide is sucrose which is present in an amount in the range from about 0.5 to about 5.0% by weight, the cyclic oligosaccharide is hydroxypropyl-.beta.-cyclodextrin which is present in an amount in the range from about 1.0 to about 10.0% by weight, all percentages by weight being based on the weight of the entire composition.

25. The cream formulation of claim 22, wherein the protease inhibitor is a serine protease inhibitor or a metalloprotease inhibitor.

26. The cream formulation of claim 25, wherein the protease inhibitor is 1.5 to 45 inh U % by weight aprotinin, 0.01 to 1% by weight EDTA, or a combination thereof, all percentages by weight being based on the weight of the entire composition.

27. An aqueous gel formulation for healing chronic wounds, comprising:
a) an effective wound healing amount of a wound healing promoter;
b) a water soluble, pharmaceutically acceptable carrier having a viscosity within the range of about 50,000 to about 1,000,000 cps at room temperature, wherein the formulation has an Abs value greater than 7.8, the Abs value being the percentage of labelled wound healing promoter present in a deepithelialized dermis sample after 12 hours in a diffusion cell system.

28. The formulation for topical wound healing of claim 27, wherein the formulation has an Abs value of at least about 13.40%.

29. The formulation for topical wound healing of claim 27, further comprising an effective amount of a preservative.

30. The formulation for topical wound healing of claim 27, wherein the wound healing promoter is an extracellular matrix protein.

31. The formulation for topical wound healing of claim 30, wherein the extracellular matrix protein is fibronectin, thrombospondin, laminin, vitronectin, or fibrinogen.

32. The formulation for topical wound healing of claim 27, wherein the wound healing promoter is at least one growth factor.

33. A method of treating a chronis wound comprising applying the formulation of claim 1 to the wound two times a day.

34. A method of treating a chronic wound comprising applying the formulation of claim 17 to the wound two times a day.

35. A method of treating a topical wound comprising applying the formulation of claim 27 to the wound two times a day.

36. An aqueous gel formulation for healing chronic wounds, comprising:
a) 0.05 to 0.5% by weight of human plasma fibronectin;
b) 0.25 to 1.0% by weight of polyacrylic acid having a molecular weight of about 740,000 to about 5,000,000 and a viscosity of about 350,000 cps, viscosity being measured at room temperature.

37. An aqueous gel formulation for healing chronic wounds comprising (w/w) 0.2% fibronectin, 0.375% carbomer, and 0.1% chlorocresol.