Pharmaceutical compositions for intramuscular administration, containing proteins and substances enhancing their absorption

78191/2 PHARMACEUTICAL COMPOSITIONS FOR INTRAMUSCULAR ADMINISTRATION, CONTAINING PROTEINS AND SUBSTANCES ENHANCING THEIR ABSORPTION. ,o>vwi qinn uw’u ηιηρίΊ >“i>wn onvpi’p cp-mini ο’υκηηΰ o>b>pan SURVIVAL TECHNOLOGY, INC. c:03890 1-1090 78191/3

FIELD AND BACKGROUND OF THE INVENTION

The present invention relates to a pharmaceutical composition for intramuscular administration, which comprises at least one therapeutically active protein and at least one agent enhancing the absorption of the protein, whereby high concentrations of the injected protein are rapidly achieved, thereby permitting rapid implementation of therapeutic effects without the need for intravenous injection or infusion. In an exemplary embodiment there is provided a vehicle comprising absorption enhancement agent, in which the protein is administered, which markedly increases permeation of proteins tested, into the vascular space, leading to high blood levels of the protein within two minutes after injection, in model tests as set forth below.

The invention is based in part upon the invention set forth in U.S. Patent No. 4,661,469 (i.e. the patent which issued from U.S. patent application serial no. 708,845). and which relates to the treatment of coronary prone individuals in the throes of a suspected myocardial infarction in such a way as to minimize damage to the heart muscle and, more particularly, to improvements in such treatments enabling the same to be commenced at the earliest possible time, even before direct qualified personal care of the individual can be established. 1 78191/2

When a clot forms in a blood vessel, the body organ being supplied with blood by that blood vessel is compromised or totally deprived of blood supply. Depending on the blood vessel in which this occurs, the threat to the life of the individual is either small or very great as in the circumstances to be addressed by the material below, i.e. certain life threatening circumstances. Clot formation in a vessel is described as thrombosis. Substances which dissolve thrombi are called thrombolytic substances. When a coronary artery clot is dissolved, the resultant establishment of blood flow to the heart is called reperfusion.

Examples of life threatening or very serious clot formation in arterial vessels are cerebral thrombosis, renal thrombosis, ophthalmic artery thrombosis, and very importantly, thrombosis of a coronary artery. In approximately 85% to 90% of cases of acute myocardial infarction (coronary heart attack), a thrombus is found in the coronary artery preventing blood from flowing to the heart muscle (myocardium) and supplying it with essential oxygen and other nutrients. A consequence of a thrombus or clot forming in a coronary artery is the danger to the myocardium (heart muscle tissue that does the pumping of the blood). Heart muscle deprived of its blood supply does not die immediately but does promptly begin the process of becoming dead.

The extent of the damage which is done to the heart muscle supply is, of therefore, a function of the time during which the blood to the infarct zone is restricted by the occluding thrombus. 2 78191/2

Heretofore, the procedures undertaken to actually establish reperfusion to the infarct zone have generally been undertaken in a hospital environment or equivalent. The so- called "pre-hospital" treatment was, in general, directed toward keeping the patient alive and getting the patient into the hospital environment as soon as possible so that the treatment minimizing the heart muscle damage could be accomplished.

The treatment undertaken in the hospital environment involves certain procedures for establishing reperfusion in the infarct zone of the patient's heart. When immediate surgery was not clearly indicated, the establishment of reperfusion was accomplished by procedures which had the effect of unblocking the occlusion. The available procedures included mechanical catheterization and the administration of thrombolytic agents.

Known thrombolytic agents, such as streptokinase or urokinase required intracoronary infusion or the slow infeed of the agent within the vessel at the site of occlusion 3 4 ζ by means of a catheter. In recent years, intravenous infusion of streptokinase has been shown to be effective.

More recently a substance called tissue-type plasminogen activator or t-PA has been utilized experimentally. (The New England Journal of Medicine, March 8, 1984, Volume 310, No. 10, pages 609-613). Unlike other plasminogen activators, such as streptokinase or urokinase, t-PA—which is found in only small amounts in the body—acts specifically on clots and not on other relevant proteins in the blood, when maintained at appropriate and effective levels. A 1984 Commentary found in Biochemical Pharmacology, Volume 33, No. 12, pages 1831-1838 entitled "Coronary Thrombolysis: Pharmacological Considerations With Emphasis On Tissue-Type Plasminogen Activator (t-PA)" contains the following conclusionary statement: 20 "Selection of pharmacological agents for induction of coronary thrombolysis has been determined largely by availability. Unfortunately, both streptokinase and urokinase induce a systemic lytic state with depletion of 25 circulating fibrinogen, plasminogen, and a 2-antiplasmin, and accumulation of fibrin degradation products. All of these factors conspire to set the stage for hemorrhage with a risk of 30 serious bleeding. Intravenous administration of these agents is limited by a lower success rate, in part because the upper 5 bound of dose is constrained by the risk of inducing a severe systemic lytic state.

The probability that progress in recombinant DNA technology will lead to widespread availability of tissue-type plasminogen activiator is particularly exciting because of the clot specific properties of t-PA. For coronary thrombolysis its potential advantages include: safety and efficacy of intravenous administration of high doses; effective clot lysis without induction of a systemic lytic state; prompt implementation without the need for extensive characterization of the coagulation and fibrinolytic systems in each patient prior to and during therapy; avoidance of frank allergic reactions or variations in dose-response relation due to immune complex formation; ease of minute-by-minute adjustment of dosage and prompt termination of fibrinolysis when needed because of the short biological half-like of t-PA and the lack of induction of a systemic lytic state."

The promise attributable to t-PA administration was discussed at a news conference at a meeting of the Americal Heart Association and reported by the New York Times on November 16, 1983, in an article entitiled, "Protein Of Cancer Cells Used To Halt Coronaries." The article refers to injection of t-PA by stating the following: "The protein t-PA can simply be injected into the vein in the art of the patient seized by a myocardial infarction or heart attack, and it travels through 6 the blood to dissolve a clot, in much the same way as Draino clears up stopped plumbing."

The article further indicated under the subheading "Hopes For Future Application" that many physicians have expressed excitement about research into the use of t-PA to treat heart attacks because they hope that some day it may be used in emergency rooms and ambulances to stop heart attacks at their earliest stages before they kill or cause permanent damage. Under the "Hopes For Future Application" subheading there is also included the following paragraph: "Dr. Burton E. Sobel Of Washington

University, one of the researchers, speculated that patients might some day carry a vial with them so that the drug could be injected immediately after they felt chest pain and other early symptoms of a heart attack."

In medical parlance, a vial is a container for a quantity of liquid medicine or diluent having a rubber stopper capable of being pierced by a hypodermic needle of a syringe to enable the operator of the syringe to withdraw a predetermined dosage of the liquid from the vial. In the case of t-PA as currently used, the dosage could then be injected into the mother liquid container of an infusion assembly. The necessity to administer the drug by intravenous infusion or by intravenous injection presents a significant barrier to selfadministration from a practical view point, particularly when considering the disconcerting circumstances of the individual undergoing the symptoms of a myocardial infarction. 7

The development of an effective selfadministration procedure for t-PA sufficient to enable its utilization by a targeted coronary prone individual immediately following onset of symptoms, would materially increase the potential efficacy of t-PA as a thromobolytic agent by insuring its use at the earliest possible time often before irreversible heart muscle damage has occurred, and, at the same time, provide a treatment of the pre-hospital or pre-ambulance type which for the first time is directly effective to minimize heart muscle damage accompanying myocaridal infarction. It is an object of the present invention to provide such a self administering treatment.

It has now been found that the invention has wider application. A number of proteins with life-saving, therapeutic properties presently must be administered exclusively intravascularly to achieve therapeutic blood levels. Examples include activators of the clot-dissolving, fibrinolytic system including streptokinase and urokinase used for interrupting heart attacks in progress, and components of the coagulation system such as antihemophiliac globulin (factor VIII) to arrest bleeding in hemophiliacs. In many circumstances other proteins such as insulin used to treat diabetic coma must be administered intravenously to yield concentrations high enough to manifest the desired therapeutic effect of promptly enough to do so within the time-constraints of the medical emergency. Recently, as pointed out above, dissolution of clots in the coronary arteries giving 78191/2 rise to heart attacks has been achieved in experimental animals and patients by intravenous administration of tissue-type plasminogen activator (t-PA), a protein capable of activating the fibronolytic system at the clot surface without predisposing the patient to bleeding. Because the half-life in the circulation of t-PA is so short, continuous intravenous infusion has been required to avoid rapid disappearance of the protein from the circulation with consequently sub-therapeutic concentrations. In addition, intravenous administration has been deemed necessary because of the urgent need for rapid clot lysis to interrupt the heart attack in progress before death of a substantial amount of heart muscle. In view of the foregoing a method has been developed with the primary objective of enhancing absorption in the blood of protein, e.g. t-PA, and other large proteins (as defined hereinbelow) injected intramuscularly. The method utilizes solubilization of the protein in a vehicle enriched with an agent enhancing absorption of the protein in the blood that elicits prompt permeation of the protein into the circulating blood pool. An additional objective of the method developed is retention of the pharmacological and therapeutic properties of the protein so that the desired therapeutic effect can be elicited by intramuscular injection with enhanced absorption. A further objective is facilitation of absorption over 3θ to 60 minutes so that therapeutic blood levels can be sustained under emergency conditions. Permeation of the proteins tested into the vascular spaces led to high blood levels within 2 minutes of injection. The vehicle developed facilitates absorption of all 8 78191 /4 four proteins in the tests mentioned below and exerts no deleterious effects on the functional properties of the two proteins tested with clot lysing properties. The invention is not limited to use with the proteins employed in the specific examples but is believed applicable to other proteins as well, such as those mentioned above, e.g. insulin.

SUMMARY OF THE INVENTION

The present invention accordingly provides a pharmaceutical composition for intramuscular administration, which comprises (1) at least one therapeutically active protein selected from streptokinase, urokinase, antihemophilic globulin, insulin, bovine serum albumin, human serum albumin and the clot selective thrombolytic protein t-PA (tissue-type plasminogen activator), and (2) at least one agent enhancing the absorption of said intramuscularly administered protein and selected from lower alkylamines, di(lower alkyl)amines , phenyl(lower alkyl)amines , hydroxyphenyl(lower alkyl)amines, hydroxy(lower alkyl)amines, methoxyamine, polyalkyleneamines and glucoseoxime, and their salts with non-toxic acids.

Where reference is made herein to the use of hydroxylamine or its salts as an absorption enhancing agent, such are used as model compounds only, but their use is not in accordance with the present invention. Persons skilled in the art will be able to substitute any of the absorption enhancing agents recited in accordance with the 9 78191/2 present invention, for hydroxylamine or its salts, without the application of any inventive skill, in accordance with the present teaching.

DETAILED DESCRIPTION OF THE INVENTION

It will be appreciated that the pharmaceutical composition according to the present invention may be applied in one embodiment for enhancing the absorption of t-PA in the blood.

There can be used for this purpose (by way of example) a known emergency type automatic injector, by means of which injection is effected into the muscle tissue, e.g. after having received a decision to do so over the telephone from a qualified source and at a time prior to the establishment of direct contact qualified personal care. t-PA is a large protein. It would not be expected that it would be absorbed into the blood stream in discernible quantities. Extra-vascular levels of protein are about 1/10 that of intravascular protein. It is thought that this is so because the capillary pores through which transport of protein can occur are small relative to the molecular size of protein and limit protein transport because of electrical charge.

It was thus highly problematical as to whether a large protein such as t-PA, when given intramuscularly, i.e. outside the blood vessels, would find its way rapidly into the blood stream in discernible quantities. Application tests have indeed shown that by itself t-PA does not find its way rapidly into the blood stream in therapeutically significant quantities after intramuscular injection. The actual treatment of the system must therefore include intramuscular injection of an absorption 1Θ 78191/3 enhancing agent simultaneously (or substantially simultaneously) with the protein, e.g. t-PA.

Augmentation of absorption of low molecular weight substances administered topically, subcutaneously, or intramuscularly has been achieved with vehicles such as dimethylsulfoxide (DMSO) and by enhancement of skeletal muscle blood and lymph flow. However, DMSO has proven ineffective as an absorption enhancing agent for t-PA and other protein molecules.

In accordance with the principles of the present invention, the absorption rate of t-PA and other proteins in the blood is enhanced by utilizing with the t-PA or other protein dosage, a dosage of an absorption enhancing agent for t-PA or other protein. The absorption enhancing agent is mixed in with the t-PA or other protein dosage to form a single mixed dosage which is then injected intramuscularly (i.m.); e.g., an amount of absorption, enhancing agent which Is mixed with the t-PA or other protein dosage, to form a single mixed dosage is an amount of from 0.1 to 85, e.g. 0.1 to 40 or 1 to 85 mg/kg body weight. 11 78191/4

In accordance with the invention, there can be used as absorption enhancing agents lower alkylamines and di(lower alkyl)amines, e.g. methylamine, dimethylamine, ethylamine, diethylamine, isopropylamine, sec.butylamine, diisopropylamine, propylamine and n-butylamine, phenyl(lower alkyl famines e.g. phenylethylamine, hydroxyphenyl(lower alkyl)amines, e.g. epinephrine and tyramine, hydroxy(lower alkyl)amines, e.g. ethanolamine, diethanolamine, triethanolamine, propanolamine, as well as methoxyamine, polyalkyleneamines e.g. ethylenediamine and diethylenetriamine. These amines can all be used in the form of salts of non-toxic acids such as salts of the acids mentioned earlier, e.g. as the hydrochlorides. Also there can be used glucoseoxime. Methylamine and dimethylamine also have the advantage that they do not induce methemoglobinemia. This was 12 78191/3 shown by injecting rabbits with (a) 0.63 molar methylamine hydrochloride and (b) 0.63 molar methylamine hydrochloride together with t-PA in an amount of 1 mg/kg body weight. In each case 1 ml of injectate was employed and the solutions had a pH of 6.5· When using methylamine (as the hydrochloride) t-PA levels were in the range of 220 to 230 ng/ml. With dimethylamine the levels of t-PA were somewhat lower but still elevated compared to the case when t-PA was administered without an absorption enhancing agent. Methoxyamine (as the hydrochloride) under the same conditions also enhanced the level of t-PA but produced methemoglobinemia. The time course of elevation of t-PA over the 30 minutes with methylamine hydrochloride was virtually the same as with hydroxylamine hydrochloride. The levels were more than 10 times the plasma t-PA levels in the absence of the absorption enhancing agent.

Also while the simultaneous administration of t-PA or the other proteins specified herein, is primarily intended for human use, the present compositions may be administered to other mammals, e.g. dogs, cats, cattle and horses. In accordance with the earlier teachings electrical stimulation of the muscle at the injection site can be employed in concert with the inclusion of an absorption-enhancing agent, e.g. hydroxylamine hydrochloride (which as previously explained is used herein as a model compound but is not per se in accordance with the invention), in the injectate in a number of the examples below using intramuscular injection. Electrical stimulation augments and enhances the absorption of the absorption enhancing agent of the invention. 13 14

Although as pointed out in the earliest parent application an automatic injector device suitable for intramuscular self-administration of t-PA or other protein can be employed, the examples set forth below were performed by administering the t-PA or other protein and hydroxylamine hydrochloride directly into the muscle with a conventional needle and syringe. Administration of the agent with an automatic injector, however, it is believed will lead to even higher blood levels than those obtainable by manual injection.

After an approach employing intramuscular injection of t-PA with hydroxylamine (as the hydrochloride) and electrical stimulation of skeletal muscle at the injection site in rabbits had been found to yield peak blood levels of t-PA comparable to or exceeding those known to elicit coronary thrombolysis after intravenous infusion of t-PA in dogs and in patients, an analogous approach was evaluated in dogs subjected to coronary thrombosis. Facilitated absorption of t-PA after intramuscular injection was found to elicit coronary thrombolysis as well as therapeutic blood levels of t-PA in these feasibility experiments.

Large injectate volumes were employed because of the limited solubility of t-PA in conventional buffers. For consistency the volumes used in rabbits were selected to be similar to those planned for use in dogs (1 and 1.5 ml per injection site for rabbits and dogs respectively) even though they represented large volumes with respect to rabbit muscle mass. Thus the same concentration of absorption-enhancing agent per ml of injectate was 15 used in both species even though they resulted in administration of markedly greater amounts of hydroxylamine per kg of body weight and a 10-fold lower concentration of t-PA in the injectates in rabbits compared with dogs despite administration of comparable proportion of t-PA administered per kg of body weight in the two species. Concentrating the t-PA appreciably with solubilizing agents such as thiocyanate it is believed will permit the volumes to be reduced substantially.

For studies in rabbits, the t-PA employed was either harvested from melanoma cell supernatant fractions (mt-Pa) as previously described (Bergmann, Science 220 1181-1183 (1983) or produced by recombinant DNA technology, Van der Werf,

Circulation 69 605-610 (1984) (rt-PA, Genentech Corp., lot BH004 DAX). Results with the two preparations were indistinguishable and therefore the preparations were pooled. Concentrations of 0.5 mg t-PA per ml buffer (0.3 M NaCl, 0.01% Tween 80, 0.01 M potassium phosphate buffer pH 7.5) were used. For studies in dogs, rt-PA (Genentech, lot TE031A) was concentrated 20-fold with an Amicon membrane filter system. DMSO was used in 1% or 3% (v/v) solutions in vitro and in injectates. Hydroxylamine hydrochloride was used in concentrations of 43.75 mg per ml of t-PA solution. This concentration was compatible with a total hydroxylamine hydrochloride dose of approximately 13 mg/kg shown to be well tolerated physiologically. 16

To determine the extent to which the absorption-enhancing agents evaluated might interact with t-PA, solutions of rt-PA (0.015 to 50 ng/ml) were incubated at 37°C for 1 hour after addition of 1% DMSO, 3% DMSO, 175 mg/ml hydroxylamine (as the hydrochloride), or both DMSO and hydroxylamine (as the hydrochloride). No effects were discernible on t-PA assayed innumoradiometrically or functionally.

Studies were performed in 56 nonfasted, white male New Zealand rabbits weighing approximately 2 kg. Endogenous t-PA in these animals does not react with antibody prepared against human t-PA and hence does not interfere with the immunoradiometric assay used to characterize blood levels of exogenously administered t-PA. Animals were anesthetized with sodium pentobarbital (24 mg/kg) and ventilated with 95% oxygen administered through a tracheostomy at 2 1/min. Skeletal muscle (vastus medialis) at the injection site was exposed bilaterally and serial blood samples were drawn through an indwelling femoral venous catheter. To augment skeletal muscle blood and lymph flow at the injection site, the muscle was stimulated for 2.0 msec at 14 volts with five pulses per second with two 27-gauge, 0.5 inch stainless steel needles. A single negative distal electrode was used as well. A total of 1 mg of t-PA/kg body weight was injected manually divided in 1 ml aliquots in each of 4 sites.

Coronary thrombosis was induced in fasted anesthetized dogs weighing approximately 23 kg, see Bergmann Science 220. 1181-1183 (1983). Occlusive thrombus formed within five to 10 minutes and was 17 78191/2 confirmed angiographically. Serial venous blood samples were obtained through an indwelling inferior vena caval catheter. Electrical field stimulation at the injection site was implemented with three 27-gauge stainless steel, one serving as the negative reference. Parameters were the same as those used in rabbits. t-PA was injected directly into exposed sartorius muscle in 1.5 ml aliquots per site such that the total dose was 3 mg/kg body weight and the total volume of injectate was 6 ml in aggregate for each dog.

The primary endpoint for experiments in the 56 rabbits studies was t-PA activity in blood. t-PA antigen levels were assayed serially as previously described Bergman, loc. cit. and Van der Werf, New Engl. Ji. Med. 310, 609-613 (1984). Functional t-PA activity was determined as well Bergman, loc. cit and Tiefenbrunn, Circulation 71, 110-116 (1985). Blood samples were obtained at 0 to 4°C in sodium citrate vacutainer tubes before intramuscular injection of t-PA or vehicle alone, immediately after injection, and at selected intervals from one to 60 minutes subsequently.

For the feasability experiments in dogs, an additional endpoint was coronary thrombolysis documented angiographically. Blood pressure, heart rate, the electrocardiogram, arterial blood gases and pH, hemoglobin and hemoglobin oxygen saturation were monitored.

For experiments in both species, a crude assessment of potential muscle injury at the site of injections was made by gross inspection. In i 78191/2 18 - addition, serial blood samples were assayed for plasma creatine kinase (CK) activity spectrophotometrically, Klein, Cardiovasc. Res. 7, 412-418 (1973) in view of the known prompt and marked liberation of CK into the circulation when skeletal muscle is inured.

Serial changes in blood levels of t-PA were evaluated in 56 rabbits comprising several groups. Blood levels were assessed before and at selected intervals after intramuscular injection of buffer with or without absorption-enhancing agent alone; or t-PA in buffer, buffer with DMSO, buffer with hydroxylamine (as the hydrochloride), or buffer with DMSO and hydroxylamine (as the hydrochloride).

The same combinations were evaluated with and without concomitant electrical stimulation of muscle at the injection site throughout the blood sampling interval. Once it had been determined that hydroxylamine facilitated absorption of t-PA, experiments were performed to define the dose-response relations for absorption of t-PA as a function of the concentration of t-PA and the concentration of hydroxylamine in the injectate. Possible systemic effects of hydroxylamine on absorption of t-PA were assessed in rabbits by administering hydroxylamine without t-PA in two injection sites and t-PA without hydroxylamine in the other two sites.

The experiments performed in dogs were undertaken after it had been determined with rabbits therapeutic blood levels could be induced with amounts of t-PA/kg body weight (1 mg/kg) of the same 19 order of magnitude as those that had been used previously for intravenous administration of t-PA in patients (0.5 to .75 mg/kg). Intramuscular t-PA was administered with hydroxylamine (as the hydrochloride) within five to 45 minutes after angiographic documentation of formation of an occlusive clot in the left anterior descending coronary artery, generally occurring within seven to 10 minutes after introduction of the thrombogenic coil into the vessel. Serial aniography was performed at approximately 15 minute intervals. Effects of t-PA on coronary thrombi correlated with plasma t-PA levels. After clot lysis (approximately 15 minutes after injection of t-PA), heparin (500 U/kg body weight) was given to prevent reocclusion. In the absence of exogenous activation of the fibrinolytic system, clots induced by the indwelling thrombogenic coronary arterial coil invariably persist despite administration of heparin (n = 40 dogs).Statistical comparisons were performed by analysis of variance with Bonferroni critical limits or with Students t test for paired data. Values are expressed as means + SE.

Effects of Absorption-Enhancing Media on t-PA Activity in vitro

Neither hydroxylamine (as the hydrochloride) (175 mg/ml), 1% DMSO, 3% DMSO, nor concomitant hydroxylamine (as the hydrochloride) and DMSO modified immunoradiometrically detectable t-PA or functionally detectable t-PA activity in samples incubated for 1 hour at 37°C containing 0.015 to 50 ng rt-PA. 20

Concentrations of t-PA in Blood

Prior to intramuscular injection of rt-PA, no human t-PA was detectable by immunoradiometric assay in plasma from any of the rabbits. No detectable endogenous t-PA activity was evident in plasma samples assayed with the fibrin plate functional assay despite the minor surgical procedure performed and the imposed electrical stimulation of muscle for 60 minutes in any of four rabbits tested. No human t-PA was detectable after injection of any of the combinations of vehicles tested when exogenous t-PA was not included in the injectate. No immunoradiometrically detectable t-PA was present in plasma samples from sham operated dogs during a 60 minute sampling interval with or without intramuscular injection of a total of 262 mg/ml of hydroxylamine as the hydrochloride administered in multiple sites. Fibrin plate assayable functional activity in sham operated dogs ranged from 10 to 53 IU/ml and did not increase in any of four animals tested during the 60 minute sampling interval after electrical stimulation and intramuscular injection of hydroxylamine hydrochloride in buffer without t-PA.

In control experiments with hydroxylamine hydrochloride alone (262 mg) injected intramuscularly in dogs, peak methemoglobin levels ranged from 11 to 13% and occurred within five to 15 minutes after intramuscular injection (n = 3). Arterial oxygen tension decreased to a minimum of 93 mm Hg. Hemoglobin saturation with oxygen declined to a minimum of 81%. Except for transitory acceleration of heart rate, dogs given hydroxylamine 21 hydrochloride with or without t-PA exhibited no significant hemodynamic or electrocardiorgraphic abnormaliites.

In the Drawings:

Figure 1 is a graph of immunoradiometrically detectable and functionally active plasma t-PA activity in plasma samples from a rabbit injected with 2 mg t-PA buffer with 43.75 mg/ml hydroxylamine hydrochloride (total injectate volume = 4 ml divided among 4 sites) followed by electrical stimulation at the injection sites throughout the sampling interval. Both immuno-reactive and functionally active t-PA peaked rapidly after intramuscular injection with facilitated absorption.

Figure 2 is a graph showing the dependence of the peak concentration plasma of immunoradiometrically detectable t-PA on the concentration of hydroxylamine in the injectate. Conditions were the same as those indicated in the legend to figure 1 except that the amounts of hydroxylamine hydrochloride in the 4 ml aggregate volume of injectate were varied as indicated in the figure.

Figure 3 is a chart showing peak plasma t-PA activity as a function of the amount of t-PA administered intramuscularly in 6 rabbits.

Conditions were the same as those indicated in the legend to figure 1 except that the total amount of t-PA administered was varied as indicated. Panel A depicts immunoradiometrically detectable activity; panel B depicts amidolytic, functional activity.

Dose related differences throughout the 1 hour interval of measurement for the entire time -activity aerol (n = 30 determinations) were 78191/2 22 significant as determined analysis of variance (p <.001).

Figure 4 is a graph showing early changes in plasma t-PA concentrations after facilitated absorption of intramuscularly administered t-PA in each of three rabbits. Conditions were the same as those indicated in the legend to figure 1.

Figure 5 a graph of serial changes in plasma t-PA assayed immunoradiometrically in a dog which had been subjected to coronary thrombosis. Thrombosis was induced with a thrombogenic coil advanced into the left anterior descending coronary artery at the tip of a coronary arterial catheter. Coronary thrombolysis was induced by facilitated absorption of intramuscularly administered t-PA. (The thrombogenic coil elicited formation of a clot evident by lack of distal fill with angiographic dye as well as by lack of opacification of the vessel proximal to the coil that appears as a bright rectangle.) Fifteen minutes after intramuscular administration of t-PA (3 mg/kg in a total injectate volume of 6 ml divided among four sites) and electrical stimulation of muscle at the injection site, lysis of the clot proximal and distal to the coil was evident with angiographically demonstrable restoration of patency. As can be seen, plasma t-PA activity peaked soon after facilitated absorption of intramuscularly administered t-PA. Elevated levels persisted throughout the sampling interval. A secondary peak was seen in each of the three dogs studied.

Figure 6 is a graph showing the effect of administering hydroxylamine hydrochloride together with bovine serum albumin. Administering the bovine 78191/2 23 serumalbumin in the absence of hydroxylamine hydrochloride. The bovine serum albumin was irradiated with iodine 125. In a first experiment it was injected intramuscularly into rabbits.

Samples of plasma were drawnout of the rabiits from time to time and the counts per minute (cpm) of radioactive iodine determined. On the graph this is identified as the curve -HA. In a second experiment the same amuont of the bovine serum albumin irradiated with iodine was injected intramuscularly into rabbits while simultaneously intramuscularly injecting hydroxylamine hydrocyloride. Samples of plasma were drawn out from time to time ad the counts per minute of radioactive iodine determined. On the graph this is identified as the curve +HA. It can be seen from the graph that in the presence of hydroxylamine hydrochloride the bovine serum albumin was absorbed more rapidly and had a sustained peak of absorption after 30 minutes of over 5 times that of the bovine serum albumin alone.

Blood Levels of t-PA After Intramuscular

Injection In Rabbits

Judging from results in experimental animals and patients given t-PA by continuous intravenous infusion, therapetuic effects are not elicited unless the concentration in blood exceeds 50 mg/ml of plasma.

As shown in Table 1, t-PA injected in buffer alone increased blood levels only minutely. The addition of DMSO to the injectate did not increase t-PA levels in plasma. In contrast, 24 hydroxylamine hydrochloride augmented absorption of t-PA yielding peak blood levels five minutes after injection approximately 40-fold higher or even more than 50 fold higher than those seen in its absence. There was noted absorption of approximately 10% of aministered protein within 30 minutes after intramuscular injection with enhanced absorption. Levels within the therapeutic range persistend throughout the 1 hour observation range (Table 1). An example of serial changes of immunoradiometrically and functional t-PA activity assayed with fibrin plates after intramuscular absorption of t-PA facilitated by inclusion of hydroxylamine hydrochloride in the injectate and electrical stimulation of muscle at the injection site is shown in Figure 1.

Table 1

Immunoradiometrically Detectable t-PA In Plasma (ng/ml) After Intramuscularly Administered t-PA t-PA in

Interval After Injection (min) t-PA In Buffer Alone t-Pa in Buffer + 3% DMSO Buffer + Hydroxylamine Hydrochloride ‘ = 6) (n = 5) (n = 15) 0 0 + 0 0 + 0 0 ί 0 5 8 + 2 11 + 4 431 + 52* 15 9 + 2 8 + 2 146 + 16* 30 9 + 2 9 + 1 85 ± 17* 60 10 + 3 10 + 1 53 + 11*

Values are means + SE. All injectates contained 2 mg t-PA in an aggregate of 4 ml (1 ml per site). The concentration of hydroxylamine hydrochloride was 43.75 mg/ml. All experiments 25 tabulated were performed with electrical stimulation of muscle at the infarction site.

* = P < .01 compared with t-PA in buffer alone or in buffer + DMSO

To determine whether augmentation of muscle blood flow by electrical stimulation would enhance absorption of t-PA administered intramuscularly, experiments were performed with and without electrical stimulation after injection of t-PA in buffer alone, t-PA in buffer supplemented with DMSO, and t-PA in buffer supplemented with hydroxylamine hydrochloride. The very low blood levels seen when t-PA was administered without hydroxylamine hydrochloride were not consistently modified by electrical stimulation (n = 11 animals). However, in animals given t-PA with hydroxylamine hydrochloride (n = 15) stimulation augmented peak levels by an average of 258 + 32% without altering the time course of absorption or clearance of t-PA.

As shown in Figure 2, immunoradiometrically detectable t-PA peak blood levels were proportional to the amount of hydroxylamine hydrochloride in the injectate. Addition of 1% or 3% DMSO to hydroxylamine (as the hydrochloride) - enriched injectates did not increase peak blood levels of t- PA compared with results with hydroxylamine hydrochloride alone when the amount of t-PA was held constant. Both immunoradiometrically detectable and functionally active t-PA after administration of exogenous t-PA were proportional to the concentration of t-PA over a four-fold range when the amount and concentration of hydroxylamine hydrochloride in the injectate were held constant 26 (Figure 3). As can be seen in Figure 4, blood levels rose rapidly and peaked between 4 and 5 minutes after injection. Appreciable concentrations of t-PA in plasma were evident as early as one minute after intramuscular injection in each case.

The augmentation of peak plasma t-PA after facilitated absorption with hydroxylamine hydrochloride was not caused simply by the decreased pH of the injectate. In each of two animals, the pH of the injectate was titrated to 5.9 without hydroxylamine. Plasma t-PA concentration five minutes after injection was only 6 ng/ml. No significant increase occurred subsequently. The increment seen with hydroxylamine hydrochloride was not attributable simply to systemic effects of hydroxylamine hydrochloride. In two animals in which hydroxylamine was injected into the right and t-PA in buffer into the left thigh muscle, peak blood levels did not exceed those in Table 1 for t-PA injected in buffer alone.

Although the amounts of absorptionenhancing agent per kg body weight used in rabbits were considerably greater than those used in dogs or anticipated ultimately for possible clinical studies, the excessively large quantities were employed to determine whether high concentrations in the injectate would be deleterious to skeletal muscle. In rabbits, plasma CK was not significantly different 30 minutes after the surgical procedure, injection of t-PA with hydroxylamine hydrochloride and electrical stimulation compared with values after injection of buffer alone under the same conditions (690 + 82 compared with 696 + 63 IU/1). - 27 - 78191/2

In dogs given 175 mg hydroxylamine hydrochloride with or without t-PA, plasma CK increased by less than 18% of baseline at the completion of the study. No hematoma were evident by gross inspection. Light microscopy of sections from the injection site obtained two hours after injection delineated only scanty interstitial hemorrhage and inflammation.

Effects of Facilitated Absorption of

Intramuscularly Administered t-PA on

Coronary Thrombolysis in Dogs

After demonstrating that facilitated absorption of t-PA could be achieved in rabbits with hydroxylamine hydrochloride in the injectate, pilot . studies were performed in dogs to determine whether the approach developed could elicit coronary thrombolysis. Arterial blood pressure after injection of hydroxylamine hydrochloride intramuscularly with (n = 3) or without (n = 3) t-PA declined only modestly (from an average of 166/121 mm Hg to 144/104) reaching a minimum 2 minutes after injection. Heart rate increased transiently by an average of 32% peaking also 2 minutes after injection. Ventricular arrhythmias did not occur with hydroxylamine hydrochloride alone.

Intramuscularly administered t-PA (3 mg/kg) followed by electrical stimulation initiated coronary thrombolysis within 15 minutes heralded by reperfusion arrhythmias. Similar results were obtained in each of the three animals studied.

Plasma t-PA values followed a similar time course but were lower than those seen in rabbits. The differences may reflect species differences in the 70191/3 - 28 absorption or clearance of human rt-.PAor the larger ratio of injectate volume to muscle mass in rabbits. In addition, as shown in Figure 5, a secondary peak of immunoradiometrically detectable 5· t-PA occurred beginning approximately 40 minutes after the first peak in each dog compatible with late release from the skeletal muscle depot because of changes in blood flow or slow lymphatic transport of t-PA into the circulation among other possibilities. 10 Thus it has been found that therapeutic blood levels of functionally active t-PA can be achieved and that coronary thrombolysis' can be eliicted by facilitated absorption of intramuscularly injected material. Plasma activity 15 peaked within five minutes after injection and subsequently declined rapidly, consistent with the known half-life of t-PA in the circulation. The blood levels obtained were sufficient to induce coronary thrombolysis in dogs within 15 minutes 20 despite the continued presence of an indwelling, coronary, thrombogenic coil. Absorption of t-PA was enhanced by inclusion of hydroxylamine in the injectate and by augmentation of skeletal muscle blood, flow by electrical stimulation. Gross injury 25 to skeletal muscle did not occur.

Because low levels of t-PA in plasma may be adequate to induce clot lysis of nascent thrombi judging from results of studies in vitro and because the biological half-life of t-PA bound to fibrin is 30 substantially longer than the half-life of circulating t-PA, see Brommer, Thromb. Res. 21» 109-115 (1984), Tran-Thang, Blood 63 1331-1337 (1984), Bergmann, Circulation 70 II:108 (Abstract) (1984), 29 it is believed that coronary thrombolysis early after the onset of thrombosis in vivo may be obtained with lower quantites of t-PA, hydroxylamine hydrochloride, or both than those used in the examples set forth above. Reduction of the injectate volume would diminish the dose of hydroxylamine or other absorption enhancing agent required and minimize potential injury to muscle at the injection site.

To date, t-PA and other activators of the fibrinolytic system have been given only by direct injection into the blood stream. This invention provides an alternative means of administration of t-PA potentially amendable to prompt implementation by paramedical personnel or by telephonically supervised patients at high risk previously instructed in self-medication procedures.

Hydroxylamine was employed after numerous attempts with other absorption-enhancing media for other compounds failed to yield the desired results with t-PA. Its major side effect, induction of methemoglobinemia does not prohibitively limit tissue oxygenation with the doses used. If the concentration of the hydroxylamine in the injectate is the critical determinant of absorption of t-PA as appears likely judging from the present results, the total dose of hydroxylamine required in human subjects is likely to be so low that induced methemoglobinemia would be of only trivial extent even for patients with ischemic heart disease especially if the injectate volume can be reduced further by increasing the concentration of t-PA. In those cases where the methemoglobinemia accompanying 30 78191/2 use of this absorption-enhancer is deemed to be unacceptably severe, adjuvant measures such as concomitant administration of methylene blue or glutathione might be utilized to minimize or obviate the problem, see Layne, J. Pharmacol. Exp. Therap. 165, 36-44 (1969).

Methylamine is a non-toxic agent. It was employed as the hydrochloride to enhance absorption of t-PA in rabbits. Absorption was measured after intramuscular injections of 2 to 3 mg of concentrated t-PA (50 mg/ml). Methylamine hydrochloride (0.63 molar) plus electrical field stimulation elected blood levels of rt-PA within 5 minutes after intramuscular injection of rt-PA, with functional and immunologic activity similar to that achieved with hydroxylamine (as the hydrochloride), (specifically 129 vs. 137 mg/ml/mg rt-PA). With the use of methylamine hydrochloride there was neither methemoglobinemia nor hemodynamic derangements oxygen saturation remained unchanged and only modest local inflammation and interstitial hemorrhage were evident microscopically in the injection site after 48 hours. Vasodilators, hypertonic media, reduced amounts of hydroxylamine hydrochloride or rt-PA alone led to much lower blood levels of rt-PA (14, 65, 46 or 4 ng/me/mg rt-PA respectively).

Blood levels of t-PA comparable to those obtained in the present investigation induce coronary thrombolysis in experimental animals and patients without inducing a systemic lytic state predisposing to bleeding. The time course of elevation of plasma t-PA after facilitated * 78191/2 intramuscular absorption is particularly favorable because of its sharp peak. With the envisioned application of an appropriate regimen, subjects would be under direct medical care soon after self-medication with an automatic injector or treatment by relatives or paramedical personnel. Thus, as the blood levels declined promptly after intramuscularly administered t-PA had been given, intravenous infusions could be initiated along with anticoagulants or other measures taken to prevent reocclusion while definitive diagnostic information was being obtained.

The possibility that myocardial reperfusion induced by facilitated absorption of intramuscularly administered t-PA might give rise to reperfusion arrhythmias is easily managed in the setting of the catheterization laboratory or coronary care unit but can be potentially dangerous in the medically unattended patient. Thus, there is advantage in the concomitant administration of an antifibrillatory or anti-arrhythmic agent such as lidocaine or an alpha-adrenergic blocking agent.

It has also been found that to prevent reocclusions or platelet aggregation it is desirable to either: 1. inhibit synthesis of thromboxane A (thromboxane A ) 2 with a thromboxane synthetase inhibitor, e.g. an imidazole such as 4-(2-[lH-imidazol-l-yl]ethoxy)-benzoic acid hydrochloride (dazoxiben). 2. introduce an antagonist for the receptor of the thromboxane A (thromboxane A ) such 2 31 32 as [lo, 2β (5Z), 3β (IE), 4a]-7-[3-(3-cyclohexyl-3-hydroxy-l-propenyl)-7-oxabicyclo[2.2.1]hept-2-yl]-5 heptenoic acid)(SQ 27,427) 3. introduce another inhibitor of platelet aggregation, e.g. aspirin, indomethacin, naproxin, and sulfinpyrazone.

The agent for the prevention of reocclusions or platelet aggregations could be administered simultaneously or sequentially in either order with reference to the t-PA and absorption enhancing agent, e.g. hydroxylamine hydrochloride. The agent for the prevention of reocclusions or platelet aggregations can be administered in conventional manner, e.g. intramuscularly, intravenously, or even orally.

The receptor antagonist or other agent for prevention of platelet reocclusions can be administered for example in an amount of 0.1-10 mg/kg body weight.

As stated previously the invention is also useful to enhance the absorption of other proteins with medicinal properties by administering intramuscularly. Such proteins include for example bovine serum albumin, human serum albumin and insulin. This is shown by the following experiments:

Two proteins, bovine serum albumin (BSA) and human serum albumin (HSA), were labeled with iodine-125 (125I) so that absorption into the circulation could be characterized by assay of radioactivity in trichloroacetic acid precipitates 33 of plasma. BSA and HSA proteins were labeled radioactively with iodine-125 (125I) by the Bolton and Hunter method to form a radioligand on tyrosine constituents. In order to verify the independence of the results on the nature of the radioligand used, a second procedure was employed in which a IZ^i-giycoconjugate was prepared to form a radiolabeled protein in which radiolabeled tyramine is bound to carbohydrate which is bound covalently to the protein. This type of radioligand is stable despite exposure to lysosomal enzymes or acidic conditions. Furthermore, it does not alter functional properties of the labeled protein. In the labeling procedure, dilactitol tyramine (10 nmol in 24 yl, in 0.5 M potassium phosphate, pH 7.7) was initially radioiodinated by placing the label in an iodogen (Pierce Chemical Company, Rockford,

Illinois) coated tube with Na125I (1 mCi in 10 yl,

Amersham). The reaction mixture was allowed to stand for 1 hour at room temperature with occasional vortexing. Contents of the tube were transferred to another tube containing galactose oxidase (4 units in 2 yl, Sigma) and incubated for 45 minutes at

37°C. Sodium cyanoborohydride (3 yl of a 1 M solution in potassium phosphate buffer) and 2 mg of bovine or human serum albumin (fraction V, Sigma) in 200 yl of physiological saline were added to the tube, followed by additional incubation for 45 minutes at 37°C. After incubation, 50 yl of 1 M 125 ammonium bicarbonate was added to the tube, I- dilactitol tyramine-albumin was separated from ^25I- dilactitol tyramine by Sephacryl S300 chromatography on a 0.9 x 30.0 cm column and dialysis against 34 physiological saline. Coupled protein was concentrated with Aquacide and sterilized by filtration through a 0.45 jum membrane filter. An aliquot was removed for determination of protein content/ specific radioactivity/ and precipitation by trichloroacetic acid.

Quantification of Absorption of Proteins after Intramuscular Injection with or without the

Absroption Enhancing Agent; Radiolabeled BSA and HSA in plasma were quantified by assay of radioactivity in a well counter of aliquots of plasma obtained at selected intervals after injection and in precipitates formed by acidification of the plasma with four parts 10% trichloroacetic acid (TCA). Enhancement of absorption of BSA and HSA proteins after intramuscular injection was demonstrated in New Zealand rabbits. Protein solutions were injected manually through a 25 guage needle into skeletal muscle in the thigh, exposed while the animals were lightly anethetized.

Enhancement of Absorption; Radiolabeled albumin was administered to rabbits in 1 ml aliquots with 163.4 to 289.9 million counts per minute associated with microgram quantities of protein (specific radioactivity = 1.74 to 3.02 mCi/mg). Proteins were administered in either 0.9% NaCl solutions or an absorption enhancing solution of 0.9% NaCl supplemented with hydroxylamine as the hydrochloride in concentrations of 43.75 to 175 mg/ml..

Enhancement of absorption of protein was 35 demonstrable with radiolabeled bovine serum albumin as well (24 determinations). Peak radioactivity in TCA precipitates from plasma averaged only 3204 counts per minutes per ml when the protein was injected intramuscularly in rabbits in saline alone. In contrast, TCA precipitable radioactivity averaged 16.580 counts per minute per ml of plasma with 80% of peak values attained 5 minutes after intramusuclar injection with absorption enhanced by hydroxylamine hydrochloride. Similarly, after injection of radiolabeled human serum albumin, TCA precipitable radioactivity averaged only 4176 counts per ml of plasma during the first 30 minutes after intramusuclar injection without enhancement of absorption. In contrast, TCA precipitable radioactivity averaged 32,868 counts per minute per ml of plasma during the first 30 minutes after intramuscular injection with absorption enhanced by inclusion of 43.8 mg of hydroxylamine as the hydrochloride per ml of injectate (12 determinations). Furthermore, 75% of peak radioactivity was evident 5 minutes after injection. The subsequent increment of plasma radioactivity after the first 5 minutes was attributable to continued absorption of radiolabeled albumin at a rate more rapid than its slow rate of clearance from the circulation.

Safety of the Absorption Enhancing Agent: Administration of hydroxylamine hydrochloride intramuscularly did not elicit gross or microscopic evidence of irreversible muscle injury at the injection site or elevation of plasma creatine kinase, a marker of muscle injury beyond the elevation seen when saline alone was used as the 36 injectate in rabbits. In dogs injected intramuscularly with 175 mg hydroxylamine hydrochloride in saline with or without protein, plasma creatine kinase increased by less than 18% above baseline and 5 no hematoma were evident at the injection site by gross inspection. Light microscopy of sections obtained from the injection site 2 hours after injection revealed only modest interstitial inflammation. Administration of the absorption 10 enhancing agent to dogs elicited only a modest decrease in systemic arterial blood pressure (from 166/121 mm Hg to 144/104) with maximal effects observed 2 minutes after injection and return to baseline within the next several minutes. Heart 15 rate increased transiently by an average of 32% with the maximal effect 2 minutes after injection and a rapid return to baseline. Hydroxylamine is known to induce methemoglobinemia. However, in the amounts utilized for enhancement of absorption of protein it 20 did not compromise oxygen carriage in the blood.

Methemoglobin peaked at 8% 2 minutes after injection in dogs, and arterial oxygen tension did not fall by more than 7%.

The invention is particularly suitable for 25 administering proteins with short biological half- lives. Thus maintenance for 30 to 60 minutes of therapeutically effective concentrations of proteins with short biological half-lives administered by intramuscular injection with enhanced absorption by 30 utilizing hydroxylamine hydrochloride was shown by induction of coronary thrombolysis with enhanced absorption of intramuscularly administered t-PA which has a half-life in the circulation of 5 to 8 minutes. 78191/1

While certain embodiments of the invention have been particularly described, it will be apparent to persons skilled in the art that many variations and modifications may be made, including obvious chemical equivalents of specified substances.

For example, while extensive reference has been made herein to the clot selective thrombolytic protein t-PA, it will be evident to skilled persons that chemically equivalent clot selective thrombolytic proteins may be substituted in whole or in part for the t-PA which has been particularly described. The invention is accordingly not to be construed as limited to the embodiments which have been particularly described, rather the concept, spirit and scope of the invention are to be construed with reference to the claims which follow. - 36a -