Immunomodulating compositions from bile

Abstract

The present invention relates to a composition for use as an immunomodulator comprising small molecular weight components of less than 3000 daltons, and having the following properties: a) is extractable from bile of animals; b) is capable of stimulating monocytes and macrophages in vitro; c) is capable of modulating tumor necrosis factor production; d) contains no measurable IL-1a, IL-1b, TNF, IL-6, IL-8, IL-4, GM-CSF or IFN-gamma; e) has an anti-proliferative effect in a malignant mouse hybridoma cell line; f) shows no cytotoxicity to human peripheral blood mononuclear cells; and g) is not an endotoxin. The invention also relates to a method of preparing the composition and its use an immunomodulator.

Description

FIELD OF THE INVENTION

The present invention relates to immunomodulating compositions, pharmaceutical agents containing the compositions, and the use of the compositions and agents in the treatment of animals.

BACKGROUND OF THE INVENTION

Therapies are continuously being developed for the prophylaxis and treatment of cancer and infectious diseases, such as Acquired Immunodeficiency Syndrome (AIDS). Some of these therapies attempt to use the immune system therapeutically. One approach is based on the antigen specific elements of the immune system, namely antibodies and T cells. For example, research has been aimed at developing vaccines against foreign agents, or against certain endogenous chemical messengers, such as interleukins, to suppress antibody reactions. A second approach is based on the isolation, cloning, expression and production of peptides and proteins from the non-antigen specific parts of the immune system. For example, proteins, such as cytokines, which comprise the interleukins produced by white blood cells, and interferons which stimulate lymphocytes and scavengers cells that digest foreign antigens, offer possibilities for therapies.

The treatment of cancer could be greatly enhanced if the early immune response to a tumor could be augmented so that the tumor does not reach a critical size. Strategies which have been suggested to augment the immune response to a tumor include vaccines specific for tumor-associated antigens; the use of monoclonal antibodies against antigens on the surface of tumor cells such as against the interleukin-2 receptor; the use of bispecific molecules containing antitumor antibodies and superantigens.

Relatively recently, the role of the physiologically active polypeptide, known as tumor necrosis factor (“TNF”) has been studied, particularly with respect to its ability to induce necrosis of tumors, with no effect upon the normal tissues of the living body. The amino acid sequence of TNF, as well as the base sequence of the DNA coding for TNF has been disclosed in U.S. Pat. No. 4,879,226.

Because TNF has been shown to have a role in inducing necrosis of tumors, any agent that can stimulate the production or bioavailability of TNF in vivo has potential utility as a treatment for various tumorous conditions. Additionally, any agent that can stimulate human monocytes and macrophages to produce TNF in vitro, is useful as a means for providing a source of TNF for therapeutic administration, as well as for analytical and diagnostic purposes.

Bile, which is secreted by the liver and stored in the gall bladder, has been investigated for various purposes, including the use of bile extracts to enhance bioavailability of drugs that are readily metabolized by normal liver function (see WO 90/12583) and to inhibit leucocytosis promotion in a mammal (see Shinoda et al., Chem. Pharm. Bull., 30, 4429-4434 (1982)). However, bile has never been considered to be a source of therapeutically useful compositions with respect to neoplastic or infectious diseases. Interestingly, in accordance with British Patent No. 337,797, it was suggested to use the gall bladder itself as a potential source of anti-cancer agents, but only after the bile had been removed from the gall bladder, and the gall bladder thoroughly washed.

SUMMARY OF THE INVENTION

It has now been discovered that bile is an important source of a composition that can stimulate TNF production both in vitro and in vivo and is effective in treating various carcinomas, especially pancreatic cancer.

The bile composition of use is obtained by extraction of bile with a water soluble or miscible solvent. The extract so obtained may be further processed to remove unnecessary or undesirable components therefrom.

The product obtained by the process of extracting bile disclosed in further detail hereinbelow has been found to have TNF stimulating activity and is believed to have anti-cancer activity, especially against pancreatic and other cancers. Obviously, the entire composition so obtained may not be necessary, to obtain such activity. Accordingly, it is possible to further separate, fractionate, or otherwise process the product thus obtained, and still retain the desired TNF stimulatory and anti-cancer activity. Moreover, it is envisioned that it is possible to obtain synthetically a product with the same or similar TNF stimulatory and anti-cancer activity. Thus, it is envisioned that the components of the product may be analyzed as to the components, or combination thereof, that are responsible for the desired activity, and a synthetic product made, based on such analysis.

In one aspect, the present invention relates to a composition for use as an immunomodulator comprising small molecular weight components of less than 3000 daltons, and having one or more of the following properties:

a) is extractable from bile of animals;

b) is capable of stimulating monocytes and macrophages in vitro;

c) is capable of modulating tumor necrosis factor production;

d) contains no measurable level of IL-1α, IL-1β, TNF, IL-6, IL-8, IL-4, GN-CSF or IFN-gamma;

e) has an anti-proliferative effect in a malignant mouse hybridoma cell line;

f) shows no cytotoxicity to human peripheral blood mononuclear cells; and

g) is not an endotoxin.

In accordance with a preferred embodiment the composition is extracted from the bile of bovines and is capable of stimulating the release of tumor necrosis factor.

The composition of the invention may be prepared by (a) mixing bile from an animal, preferably a bovine, with a solvent that is soluble or miscible with water, preferably an alcohol, and preferably with an equal volume of an alcohol, to produce a bile/alcohol solution; (b) separating the solution which preferably is an alcohol soluble fraction, and isolating therefrom a solution substantially free of alcohol, as by removing most of the alcohol, such as by the use of heat; (c) removing bile pigments from the solution to obtain a colorless liquid; (d) optionally treating the colorless liquid to substantially remove any residual alcohol; (e) removing fatty organic materials, as by extracting the colorless liquid with ether and isolating the aqueous phase; and (f) optionally removing residual ether from the aqueous phase.

The composition may be used without further modification by simply packaging it in vials and sterilizing. The composition may be also be used in a concentrated form. A preferred concentrated form is prepared as follows. Prior to step (e) the colorless liquid may optionally be concentrated to about one eighth of the volume of the bile/alcohol solution and after step (f) the aqueous phase may be concentrated so that it is one tenth of the volume of the bile/ethanol solution.

The invention also relates to a pharmaceutical agent comprising the novel composition of the invention.

The invention further relates to a method of treating a patient comprising administering to said patient an effective amount of a composition of the invention. The invention still further relates to the use of a composition of the invention in the prophylaxis and treatment of diseases and conditions requiring modulation of the immune response; preferably infectious diseases and neoplasias.

These and other aspects of the present invention will become evident upon reference to the following detailed description and attached drawings. In addition, reference is made herein to various publications, which are hereby incorporated by reference in their entirety.

BRIEF DESCRIPTION OF THE DRAWINGS

Further details of the invention are described below with the help of the examples illustrated in the accompanying drawings in which:

FIG. 1 is an HPLC profile for a concentrated composition of the invention;

FIG. 2 is an HPLC profile for a concentrated composition of the invention;

FIG. 3 is an HPLC profile for a concentrated composition of the invention;

FIG. 4 is an HPLC profile for a composition of the invention;

FIG. 5 is an HPLC profile for a composition of the invention;

FIG. 6 is an HPLC profile for a composition of the invention;

FIG. 7 is an HPLC profile for a composition of the invention;

FIG. 8 is a graph showing the effect of the composition on LPS-induced release of TNF by PBMN;

FIG. 9 is a bar graph showing the effect of the composition on LPS-induced release of TNF by PBMN;

FIG. 10 shows the C 18 RP-HPLC profile of a composition of the invention;

FIG. 11 shows the RP-HPLC analysis of precipitated fractions of the composition of the invention;

FIG. 12 shows the RP-HPLC analysis of soluble fractions of the composition of the invention.

FIG. 13 is a graph showing survival taken from diagnosis of pancreatic cancer patients treated with the composition of the invention;

FIG. 14 is a graph showing survival taken from treatment of pancreatic cancer patients treated with the composition of the invention;

FIG. 15 is a graph showing survival of all melanoma patients treated with the composition of the invention;

FIG. 16 is a graph showing survival of melanoma patients with two or more tumor sites treated with the composition of the invention;

FIG. 17 is a graph showing survival of melanoma patients with three or more tumor sites treated with the composition of the invention;

FIG. 18 is a graph showing the RP-HPLC profile of whole composition of the invention;

FIG. 19 is a graph showing the RP-HPLC profile of a precipitate of the composition of the invention;

FIG. 20 is a graph showing the RP-HPLC profile of a supernatant of the composition of the invention;

FIGS. 21A and 21B show the conditions and times of elution of the composition of the invention on hydrophilic HPLC (a) and the elution profile for a supernatant of the composition of the invention (b);

FIG. 22 shows the elution of a precipitate of the composition of the invention on hydrophilic HPLC;

FIG. 23 is a graph showing dose response of the composition of the invention in stimulating peripheral blood monocyte function.

DETAILED DESCRIPTION OF THE INVENTION

As hereinbefore mentioned, the present invention relates to a composition for use as an immunomodulator comprising small molecular weight components of less than 3000 daltons, and having the following properties:

a) is extractable from bile of animals;

b) is capable of stimulating monocytes and macrophages in vitro;

c) is capable of modulating tumor necrosis factor production;

d) contains no measurable level of IL-1α, IL-1β, TNF, IL-6, IL-8, IL-4, GN-CSF or IFN-gamma;

e) has an anti-proliferative effect in a malignant mouse hybridoma cell line;

f) shows no cytotoxicity to human peripheral blood mononuclear cells; and

g) is not an endotoxin.

More particularly, investigations have shown that the at least some of the compositions of the invention will stimulate normal monocytes to effect cytotoxicity towards the Chang hepatoma cell line, which is used to measure monocyte toxicity. Monocytes and macrophages from cancer patients (cervical and ovarian cancer) also have been reported to be stimulated by the composition to attack and destroy their own particular tumor cells.

The composition of the invention can modulate tumor necrosis factor (TNF) production. A preferred composition of the invention isolated from bile from bovines, promotes the release of TNF from human peripheral blood mononuclear cells in what appears to be physiological quantities. Because TNF is known to initiate a cascade of inflammatory and antitumor cytokine effects, the preferred composition could exert its antineoplastic effect by stimulating human leucocytes to release TNF (and possibly other cytokines). Accordingly, the present invention also may enhance lymphocyte and macrophage cytotoxicity towards tumor cells.

The composition of the invention has also been found to inhibit the growth of cells of mouse hybridoma cell line #6-1. The inhibitory effect of the composition in the mouse hybridoma cells suggests antiproliferative activity.

The effect of the composition on the survival of human peripheral blood mononuclear cells (PBMN) was also examined. The composition was found to be non-cytotoxic to human PBMN.

As further exemplified below, the composition of the present invention has, among others, the following characteristics:

1) The component or components responsible for TNF-release from PBMN eluted early from a C 18 RP-HPLC column.

2) The TNF-releasing component(s) is (are) precipitated, in part, by 80% acetonitrile.

3) The material unprecipitated by 80% acetonitrile retains some TNF-releasing activity.

4) The TNF-releasing activity in both the 80% acetonitrile precipitate and supernatant fractions eluted at the same early time from RP-HPLC. The results suggest that active TNF-releasing components in the composition belong to the same molecular family with perhaps some subtle molecular differences that account for solubility differences.

5) The composition causes the release of interleukin-1β (IL-1β), and the component responsible for the IL-1β release elutes early from RP-HPLC, suggesting that it is likely the same substance(s) that releases TNF.

6) The composition also causes the release of low quantities of interleukin-2 (IL-2).

7) The composition causes the release of granulocyte macrophage colony stimulating factor (GM-CSF); the 80% acetonitrile precipitate fraction is more active than the supernatant fraction.

8) The ratio of TNF to GM-CSF release is about 2:1.

9) The 80% acetonitrile precipitate fraction contains component(s) that release about 3 fold more TNF and GM-CSF than component(s) in the supernatant fraction.

10) Analysis of the aforementioned precipitates and supernatant fractions separated by RP-HPLC shows that releasing activity for TNF, IL-1β and GM-CSF elutes early for both the precipitate and supernatant. However, in the supernatant, some IL-1β activity elutes late.

11) It is likely that the same molecule(s), i.e., component(s), in the composition are responsible for releasing TNF, IL-1β and GM-CSF. It is possible that the composition acts to stimulate the release of multiple different cytokines, or alternatively, the composition triggers the production and release of one cytokine that in turn stimulates production and release of other cytokines.

12) Physicochemical analysis of the composition, including the precipitates and supernatants thereof, by SDS gel electrophoresis and molecular sieve HPLC indicates that the principal components are less than 2500 daltons.

13) Further physicochemical separation by hydrophilic (polyhydroxyethyl) molecular sieve HPLC confirms the small molecular weight of the components in the composition.

14) Amino acid analysis before and after acid hydrolysis suggest the presence of peptide bonds, indicating the presence of peptides.

15) Amino acid content of the active fraction from RP-HPLC shows high levels of glutamate/glutamine and glycine. In addition, residues of asparagine, threonine, serine and alanine were detected.

16) There are some unidentified ninhydrin positive residues that are likely free amino acids.

As hereinbefore mentioned, the composition of the invention may be prepared by (a) mixing bile from an animal, preferably a bovine, with an equal volume of an alcohol to produce a bile/alcohol solution; (b) separating out the alcohol soluble fraction and isolating a solution substantially free of alcohol; (c) removing bile pigments from the solution to obtain a colorless liquid; (d) treating the colorless liquid to substantially remove any residual alcohol; (e) extracting the colorless liquid with ether and isolating the aqueous phase; and (f) removing residual ether from the aqueous phase.

The composition is obtained from the bile of any animal which produces bile, preferably non-human animals. While the composition may possess a different activity toward a specific disease if obtained from the bile of one species as opposed to another, a generally suitable source of bile is that taken from bovines, ovines and swine. In most cases, it is practical to obtain the bile of slaughtered healthy food animals, such as bovines, ovines and pigs, for use in the preparation of the composition of the invention. The bile thus collected should come directly from the gall bladders of the slaughtered animals and should be substantially clear, thereby indicating that the bile preparation has a low mucus content and is substantially free of pus or blood.

In a preferred embodiment of the method, bile from bovine sources is utilized. Bovine bile is plentiful, because, in part, relatively large quantities can be extracted from each animal. Moreover, bovines are routinely slaughtered and inspected under health-related regulations, thus such animals provide a reliable source for preparing the composition of the invention. Furthermore, humans are less likely to have an allergic reaction to material of bovine origin.

The bile is mixed with an equal volume of an alcohol to produce a bile/alcohol solution, which is 50% alcohol. The alcohol may be an aliphatic alcohol, preferably methanol, ethanol, or propanol, most preferably ethanol.

A solution that is substantially free of the 50% alcohol-insoluble material may be isolated by centrifuging. Preferably, the bile/alcohol mixture is centrifuged at 3000-5000 RPM, most preferably 4200 RPM, for at least 2 hours, at about 15-25° C. The alcohol contained in the bile/alcohol-soluble fraction then may be removed by taking advantage of the different volatility of alcohol and water, using conventional methods, i.e., heating the fraction to a suitable temperature, e.g., 80-85° C., for a suitable amount of time, e.g., up to about 10 hours.

Bile pigments may be removed from the solution to obtain a colorless liquid by using activated charcoal, polyamidic microgranules, or filtration. Preferably, an activated charcoal treatment is utilised. The procedure may be repeated in order that the solution satisfies optical density and conductivity standards.

The colorless liquid is treated to remove substantially any residual alcohol, using conventional methods. Preferably the colorless liquid is filtered using a filter having about a 1.0-3.5 μm retention, most preferably a retention of 2.5 μm.

The colorless liquid is then extracted with ether and the aqueous phase is isolated. The ether used in this step is preferably dimethyl ether, ethyl ether, n-propyl ether, isopropyl ether, or n-butyl ether, most preferably ethyl ether.

Residual ether may be removed from the aqueous phase by, for example, heating the solution up to 55° C., preferably up to about 40° C. for about 5-15 hours, most preferably for about 10 hours.

The composition may be used without further modification simply by packaging it in vials and sterilizing. The composition also may be used in a concentrated form. A preferred concentrated form is prepared as follows. Prior to step (e) described hereinabove, the colorless liquid optionally may be concentrated to about one eighth of the volume of the bile/alcohol solution by, for example, heating to a temperature of less than about 85° C., preferably, to about 60°-70° C. After step (f), the aqueous phase may be concentrated so that it is one tenth of the volume of the bile/ethanol solution by, for example, heating to about 80-85° C.

In a preferred method to prepare a composition of the invention, the collected bile is mixed with an equal volume of ethyl alcohol. The bile/alcohol mixture is then centrifuged at about 4200 RPM for at least 2½ hours, at about 20±2° C. The supernatant liquid is decanted and checked for pH and ethanol content. Bile pigments are then removed using activated charcoal. The treated bile/ethanol solution is then monitored for optical density (O.D.) and conductivity. O.D. levels or conductivity levels outside acceptable specifications require that the bile/ethanol solution be given additional treatment to remove bile pigments, for example treatment again with activated carbon to achieve a reading within specification limits.

Following activated carbon treatment, the solution is filtered through a filter having a 2.5 μm retention, the alcohol is evaporated off by heating to less than 85° C. and the solution is concentrated to approximately one eighth of the original bile/ethanol solution volume. The concentrated solution is cooled to between about 20-25° C. This solution is then mixed with ethyl ether and the ether phase is discarded. Preferably, relatively small volumes of ether and strong agitation are used, such as 0.1 to 1 volume, preferably 0.2 to 0.5 volume. This step may be repeated once. The aqueous phase is heated to remove residual ether by heating up to 55° C. for about 10 hours, and further reduced in volume to one tenth of the original bile/ethanol volume by heating to about 80-85° C. This solution is then tested for appearance, biological activity, and ethanol and ether content.

The pH of the composition may be adjusted to physiological pH, i.e. 7.4-7.5, using hydrochloric acid (1%) solution and sodium hydroxide (1% solution), and a buffered solution may be obtained using dibasic and monobasic sodium phosphate salts as buffers, using conventional methods.

The composition may be used without further modification by simply packaging it in vials and sterilizing. A preferred sterilization method is to subject the composition to three sterilization cycles by autoclaving followed by incubation.

The composition may be used in a concentrated form. The preparation of the concentrated form is described above. The composition may also be lyophilized.

The composition and concentrated composition are clear yellowish solutions essentially free of foreign matter, containing not more than 10 ppm ethanol and not more than 5 ppm ether. In the bioassay described in Example 4, the composition has been shown to cause non-proliferative growth at about 18 units per ml.

The compositions of the invention can be produced in a consistently reproducible form using the method as generally described above with demonstrated identity, potency and purity from batch to batch. Identity and purity are determined using reverse-phase high pressure liquid chromatography. (See Example 1). The compositions of the invention have a consistently reproducible pattern on reverse-phase HPLC, in which peaks are seen early in the exclusion fraction at about 27 and 32 minutes. Before, in-between and after the tall peaks, there are smaller peaks that vary in intensity. The HPLC readings for three lots of the concentrated composition of the invention are shown in FIGS. 1 to 3 . RP-HPLC profiles for batches BO211 (FIG. 4 ), BO209 (FIG. 5 ), B29/3006 ( FIG. 6 ) and B15/1606 (FIG. 7 ), also show a very reproducible pattern. The compositions also display non-proliferative growth of about 18 units per ml in the bioassay described in Example 4. The compositions are also characterized by the properties hereinbefore mentioned, for example their ability to stimulate monocytes and macrophages in vitro, etc.

Compounds likely to be present in the present composition, considering the source, include sulfonated bile acids, oxidized bile acids, other naturally occurring bile acids, and their amino acid (especially glycine and taurine) conjugates and sterols. Accordingly, it is believed that the present composition includes at least one compound having the formula

wherein the molecule may or may not be fully saturated, such that, for example, the bond between A and B, B and C, or C and D may be single or double bonds, and wherein X is H, OH, ═O, or OSO 3 H; and Y is

wherein R is an amino acid residue, such as, for example, glycyl, glutamyl, or tauryl, thereby forming the glycine or taurine conjugate.

In particular, the composition of the present invention has been analyzed as to its component compounds, including organic and inorganic components. Such information was derived using standard methods of analytical chemistry, including mass spectroscopy (MS). The results of such studies include, for example, the identification of specific bile acid compounds thought to be present, including cholic acid, glycocholic acid, deoxyglycocholic acid, ursodeoxycholic acid, cholesterol sulfate, deoxycholic acid, chenodeoxycholic acid, and taurocholic acid.

From the MS it is not distinguishable if the loss of OH and H 2 of some compounds are occurring in the MS or if the deoxy, dideoxy and unsaturated analogs of such compounds are also present to begin with. These compounds may all be present as salts of ammonium, aklylammonium and inorganic cations.

The MS analysis also supports the identification in the present composition of phospholipids, sphingolipids and related agents capable of forming miscelles. Specific compounds thought to be present include:

stearic acid CH 3 (CH 2 ) 16 COOH,

palmitic acid CH 3 (CH 2 ) 14 COOH

oleic acid Z-9 octadecanoic acid: CH 3 (CH 2 ) 2 CH 2 CH═CHCH 2 (CH 2 ) 6 COOH

oxidized or hydroxylated/unsaturated short chain fatty acids: C 6 H 8 O 3 (e.g., CH 3 CH═CHCOCH 2 COOH or a C 6 acid with 2 double bonds and a hydroxide)

acetic acid

stearic acid diglyceride

palmitic acid diglyceride

stearic acid, palmitic acid diglyceride

stearic acid-monoglyceride-phosphocholine (a lysolecithin)

stearic acid monoglyceride

stearic acid triglyceride

palmitic acid monoglyceride

phosphocholine

phosphoserine

phosphosphingosine

stearic acid-sphingosine

sphingosine

stearic acid amide

stearic acid methylamide

palmitic acid amide

lecithin

sialic acid-glycerol dimer

In addition, preliminary HPLC and titration evidence has been obtained which shows that shorter chain fatty acids are also present.

Phospholipid, sphingolipid, and related hydrolysis product compounds likely to be present considering the source and the information derived from the MS and HPLC analyses include at least one compound having the formula

where R′, R 2 , R 3 are different or the same and are H, COR 4 , CH═CH—R 5 , X, —P(O)(OH)O—, or —S(O) 2 —; X is selected from the group consisting of choline, ethanol amine, N-alkylated ethanolamines, serine, inositol, sugars bearing free hydroxyls, amino-sugars, sulfonated sugars, and sialic acids; R 4 is C 1 -C 30 alkyl that is saturated or unsaturated, oxidized or hydroxylated; and R 5 is an alkyl group or oxidized and/or hydroxylated analogs thereof.

The fatty acids and their conjugates may be present in the aforementioned aqueous extract as salts. The solubility of such compounds is also enhanced by other components of the mixture. Amides of the included carboxylic acids, RCONR′R 2 , where R′ and R 2 are the same or different and are H or alkyl, are also believed to be present.

A third class of compounds, namely, mucin and proteoglycan hydrolysis products, are also likely to be present, considering the source of the composition and the aforementioned MS analysis thereof. Such compounds include hydrolysis products of mucoproteins from bile and from the gallbladder wall, such as: chondroitin 4- and 6-sulfates, dermatan sulfate, heparin, heparin sulfate, hyaluronic acid and the hydrolysis products (monomers, dimers, oligomers and polymers) of these mucins. Chitin and other mucins may be similarly hydrolyzed, which hydrolysis products would include:

N-acetyl-D-glucosamine, N-acetyl-D-galactosamine-4-sulfate, galactose-6-sulfate, N-acetyl-D-glucosamine-6-sulfate, glucosamine-6-sulfate, D-glucosamine 2-sulfate, D-glucosamine 2,3-disulfate, D-galactose-6-sulfate, glucuronic aid 2-sulfate, N-acetylneuraminic acid, sialic acid, N-acetyl chondrosine, chondroitin 4-sulfate, chondroitin 6-sulfate, D-glucosamine, D-galactosamine, glucuronic acid, glucose, galactose, mannose, fucose, iduronic acid, hexose, hexosamine, ester sulfate, glucuronic acid, chondrosamine, 2-amino-2-deoxy-D-galactose, serine, proline, threonine, alanine glycine taurine, glutamic acid, aspartic acid, histidine, and small peptides.

Similar products would be obtained by hydrolysis of mucins such as keratin sulfates, dermatan sulfates the natural sugar-sugar linkages in the dimers, oligomers and polymers may be replaced by —O—Si(OH) 2 —O— bridges between the sugar monomers or adjacent sugar chains.

In particular, specific mucin and proteoglycan hydrolysis product compounds thought to be present include:

sialic acids and their mono and diacetylated and glycolylated monomers;

N-acetylneuraminic acid;

hexosamines;

L-fucose;

hexosamine-hexuronic acid (dimer) disulfate;

glucuronic acid or iduronic acid disulfate, monoacetylated;

sialic acid-glycerol (dimer); and

dimers, trimers, oligomers and polymers of the above monomers in acetylated and sulfated form.

A fourth class of compounds, namely fat-soluble vitamins, likely to be present considering the source and the aforementioned MS analysis, include A, D, and K vitamins (e.g., D1, D3, D4, K1, K2, K5, K6, K7, K-S(II), and Vitamin E acetate, for example.

In particular, specific fat-soluble vitamin compounds thought to be present include at least one of the group consisting of Vitamin A2, Vitamin D1, Lumisterol (present from its vitamin D1 complex), Vitamin E, Vitamin K1 oxide, and Vitamin K5.

Various miscellaneous organic compounds are likely to be present, considering the source and the aforementioned MS analysis, such compounds include:

bilirubin, and its gluconuride conjugate;

biliverdin, and its gluconuride conjugate;

traces of steroids;

other plasma solutes, such as sugars, purines and pyrimidines;

miscellaneous dietary lipids; and

glutathione and its hydrolysis products.

In particular, specific miscellaneous organic compounds believed to be present in the composition include at least one of the group consisting of urea, methyl amine, dimethylamine, ethylamine, methylethylamine, diethylamine, dipropylamine, butylethylamine, ammonia, choline, taurine, glutamic acid, glycine, alanine, p-ser, p-eu, p-ea, asp thr ser sar, a-aba, cit, val, ile, leu, B-ala, G-aba, OH-lys, orn, lys, butylated hydroxy toluene (BHT), and polyethylene glycol.

Amines present in the present composition, particularly the secondary amines, may include nitrogen oxides from the air, thus forming nitroso compounds. N-oxides and N-carbamate byproducts may also be included. This series of amines cited above should be extended to include all primary, secondary and tertiary alkylamines.

Certain inorganic elements have been identified and quantified (mg/l) as follows:

Tungsten   0.07
Zinc       0.666
Phosphorus 378
Cadmium    0.01
Cobalt     0.008
Nickel     0.022
Barium     0.032
Iron       0.022
Manganese  0.039
Chromium   0.060
Magnesium  7.46
Aluminum   0.136
Calcium    5.97
Copper     0.087
Titanium   0.01
Strontium  0.060
Sodium     9600
Potassium  483
Chloride   15400
Ammonia    218
Vanadium   1 ppm

The compositions of the invention have valuable pharmacological properties. In particular, the compositions of the invention have anti-proliferative effects, effect neoplastic growth, and effect release of tumor necrosis factor. The compositions have been shown to cause no significant toxicity and only transient adverse side effects (for example, slight fever, polydipsia, pain at injection site). They have also been found to contain no detectable components of high molecular weight matter (i.e., above about 5,000 daltons), which can cause harmful immunologic reactions. The compositions may be used as agents for the prophylaxis and treatment of conditions requiring modification of the immune response, in particular infectious diseases, neoplasias, and autoimmune diseases. They may be especially useful in the treatment of various forms of neoplasia, such as leukemias, lymphomas, melanomas, adenomas, sarcomas, and carcinomas. In particular, the composition may be useful for treating malignant melanoma, pancreatic cancer, cervico-uterine cancer, cancer of the kidney, stomach, lung, rectum, breast, bowel, gastric, liver, thyroid, neck, cervix, salivary gland, leg, tongue, lip, bile duct, pelvis, mediastinum, urethra, bronchogenic, bladder, esophagus and colon, and Kaposi's Sarcoma, which is a form of cancer associated with HIV-infected patients with Acquired Immune Deficiency Syndrome (AIDS). The composition may also be used for other anti-proliferative conditions, such as arthrosclerosis and viral infections, in particular AIDS. It may also be used in the treatment of autoimmune diseases, including multiple sclerosis, rheumatoid arthritis, systemic lupus erythematosus, Type I diabetes, myasthenia gravis, Addison's Disease, autoimmune hemolytic anaemia, Crohn's disease, Goodpasture's syndrome, Graves' disease, Hashimoto's thyroiditis, idiopathic thrombocytopenic purpura, pernicious anaemia, poststreptococcal glomerulonephritis, psoriasis, scleroderma, Sjogren's syndrome, spontaneous infertility, and pemphigus vulgaris.

The compositions of the invention may be converted using customary methods into pharmaceutical agents. The pharmaceutical agents contain the composition of the invention either alone or together with other active substances. Such pharmaceutical agents can be for oral, topical, rectal, parenteral, local, inhalant, or intracerebral use. They are therefore in solid or semisolid form, for example pills, tablets, creams, gelatin capsules, capsules, suppositories, soft gelatin capsules, gels, membranes, and tubelets. For parenteral and intracerebral uses, those forms for intramuscular or subcutaneous administration can be used, or forms for infusion or intravenous or intracerebral injection can be used, and can therefore be prepared as solutions of the compositions or as powders of the active compositions to be mixed with one or more pharmaceutically acceptable excipients or diluents, suitable for the aforesaid uses and with an osmolarity which is compatible with the physiological fluids. For local use, those preparations in the form of creams or ointments for topical use or in the form of sprays may be considered; for inhalant uses, preparations in the form of sprays, for example nose sprays, may be considered. Preferably, the composition is administered intramuscularly.

The pharmaceutical compositions can be prepared by per se known methods for the preparation of pharmaceutically acceptable compositions which can be administered to patients, and such that an effective quantity of the active substance is combined in a mixture with a pharmaceutically acceptable vehicle. Suitable vehicles are described, for example, in Remington's Pharmaceutical Science (Nack Publishing Company, Easton, Pa., U.S.A. 1985).

On this basis, the pharmaceutical agents include, albeit not exclusively, the composition of the invention in association with one or more pharmaceutically acceptable vehicles or diluents, and are contained in buffered solutions with a suitable pH and iso-osmotic with the physiological fluids.

The compositions are indicated as therapeutic agents either alone or in conjunction with other therapeutic agents or other forms of treatment. For example, in the case of a malignant tumor, the present treatment may render a tumor suitable for surgical removal where it was not previously operable. The compositions and agents of the invention are intended for administration to humans or animals.

In general, a dosage range of the composition is envisaged for administration in human medicine of from about 0.01 to 20 mg/kg, preferably from about 0.1 to 10 mg/kg, most preferably 0.1 to 1 mg/kg of body weight daily may be employed. In the case of intravenous administration, the dosage is about 0.1 to 5 mg/kg of body weight daily, and in the case of oral administration the dosage is about 1 to 5 mg/kg of body weight daily. Where the concentrated composition is used, approximately half the above mentioned dosages may be used. For example, for intramuscular administration, a dosage of about 0.2 to 1.0 mg/kg of body weight daily, preferably 0.275-0.75 mg/kg of body weight daily may be used.

It will be appreciated by medical practitioners that it may be necessary to deviate from the amounts mentioned and, in particular, to do so as a function of the body weight and condition of the animal to be treated, the particular disease to be treated, the nature of the administration route and the therapy desired. In addition, the type of animal and its individual behaviour towards the medicine or the nature of its formulation and the time or interval at which it is administered may also indicate use of amounts different from those mentioned. Thus it may suffice, in some cases, to manage with less than the above-mentioned minimum amounts whilst in other cases the upper limit mentioned must be exceeded. Where major amounts are administered, it may be advisable to divide these into several administrations over the course of the day.

Thus, the present invention comprises a process for preparing an immunomodulator composition comprising (a) mixing bile from an animal with a water-soluble solvent to produce a bile/solvent solution; (b) isolating an aqueous solution substantially free of solvent from the bile/solvent solution; and (c) removing bile pigments from the substantially solvent-free solution to obtain a colorless liquid, preferably where the water soluble solvent is an alcohol, and where the bile from the animal is mixed with an equal volume of the alcohol. Preferred aspects of the aforementioned process also comprise further concentrating the colorless liquid to about one-eighth, or one-tenth, the original volume of the bile/solvent solution. Obviously, compositions produced via the above process form a preferred aspect of the invention.

The present invention also comprises a composition for use as an immunomodulator, comprising at least one component having a molecular weight of less than about 3000 daltons, which shows no cytotoxicity to human peripheral blood mononuclear cells, and has at least one of the following properties:

(a) is capable of stimulating monocytes and macro-phage in vitro or in viva to produce one or more cytokines;

(b) is capable of stimulating monocytes or macro-phages to produce tumor necrosis factor in vitro or in vivo; or

(c) has an anti-proliferative effect in a malignant mouse hybridoma cell line; and

wherein said component is not an endotoxin, IL-1α, IL-1β, TNF, IL-4, IL-6, IL-8, GM-CSF or IFN-gamma. Such compositions may be obtained from the bile of animals, preferably bovines, or from other sources. In a preferred embodiment of the composition, the composition stimulates tumor necrosis factor production in vitro or in vivo, and most preferably in humans, in the absence of exogenous IL-1α, IL-1β, TNF, IL-4, IL-6, IL-8, GM-CSF, and IFN-gamma,

The compositions of the present invention also have components which can be characterized by column chromatography such that when said composition is dried to obtain a solid residue, and 2 grams of said residue are dissolved in 20 ml of a 10% concentrated ammonium hydroxide solution in methanol, and after any insoluble material is removed, is subjected to column chromatography in a methanol column having dimensions of 5 cm×12.5 cm, and containing 102 g of 60 A flash silica gel, and operating at a pressure of 10 pounds per square inch and a flow rate of 11 mi/min with a 10% concentrated ammonium hydroxide in methanol solvent solution, said component is eluted from the column in a fraction taken when the total column elution is between about 180 and about 220 ml, between about 220 ml to about 260 ml, or between about 260 ml and about 300 ml.

Characterization of components may also be accomplished by ion-exchange chromatography, such that when 10 ml of said composition is subjected to anion-exchange chromatography in a column containing Bio-Rad AG-1 hydroxide form resin in an amount sufficient to bind substantially all the anions present in said 10 ml of said composition, said component is eluted from the column using a step gradient of ammonium bicarbonate buffer at a buffer concentration from about 0.5 M to about 1.5 M, preferably at a buffer concentration from about 1.0 M to about 1.5 M, and most preferably at a buffer concentration of about 1.5 M.

Reversed-phase (C18) HPLC can also be used for characterization of components, such that when said composition is lyophilized and reconstituted in 0.1% TFA in water and then subjected to reversed-phase (C18) HPLC in a Phenomenex WP60009-C18 column, having dimensions of 250×4.6 mm, where a first buffer of 0.1% TFA in water is run through the column for about 10 minutes, then a linear gradient from 0 to 80% of a second buffer of 0 1% TFA in acetonitrile is run for about 55 minutes, followed by an 80% solution of the second buffer for about 5 minutes, and an 80%-0% gradient of the second buffer for about 5 minutes, and where flow rate is 1 ml/min. and the capacity of the column and buffers are not exceeded, said component is eluted from the column at a time from about 2.4 minutes to about 3.4 minutes after said reconstituted composition is applied to the column. Characterization of components of the composition can also be accomplished by an additional reversed-phase HPLC method, such that when said composition is dialyzed or dissolved in a first buffer of 0.1% TFA in water and then subjected to reversed-phase (C18) HPLC in a Bio-Rad Hi-Pore RP 318 (C18) column, having dimensions of 250×4.6 mm, where the first buffer is run through the column for about 10 minutes, then a linear gradient from 0-80% of a second buffer, of 0.1% TVA in acetonitrile is run for about 55 minutesB followed by an 80% solution of the second buffer for about 5 minutes, and an 80-0% gradient of the second buffer for about five minutes, and where the flow rate is 1 ml/min. and the capacity of the column and the buffers are not exceeded, said component is eluted from the column at a time from about 2 minutes to about 21.4 minutes, or at a time from about 21.4 minutes to about 25.6 minutes after said dialyzed composition is applied to the column.

The compositions of the present invention can also be characterized by TLC, such that when said composition is subjected to thin layer chromatography on silica gel plates in 10% concentrated ammonium hydroxide in methanol and visualized with a ninhydrin spray, a positive reaction with ninhydrin occurs at an R f value from about 0.80 to about 0.90.

The present invention also comprises a method of stimulating tumor necrosis factor production in humans, comprising administering an effective amount of a composition comprising at least one of the following compounds:

(a) a compound of the formula

where the bonds between A—B, B—C, and C—D may be single or double bonds, and where X═OH, ═O, or OSO 3 H; and Y═

where R is an amino acid residue;

(b) a compound of the formula (R 1 O)CH 2 CH(OR 2 )CH 2 (OR 3 —X) or

where R 1 , R 2 and R 3 are H, COR 4 , CH═CH—R 5 , X, P(O) (OH)O—, or —S(O) 2 O—;

X is choline, ethanolamine, N-alkylated ethanolamines, serine, inotitol, sugars bearing free hydroxyls, amino-sugars, sulfonated sugars, or sialic acids; and

R 4 is a saturated or unsaturated alkyl group having a carbon chain from about C 1 to C 30 , or oxidized and hydroxylated analogs thereof; and

R 5 is an alkyl group or oxidized and hydroxylated analogs thereof,

(c) a mucin hydrolysis product or a proteoglycan hydrolysis product; or

(d) a fat-soluble vitamin.

Preferably, compositions of the inventive method comprise at least one compound selected from the group consisting of taurocholic acid and its sulphated derivatives; glycocholic acid and its sulphated derivatives; sphingosine; a diacyl glycerol; lecithin; an oligosaccharide of less than 10 saccharide units in length, where said oligosaccharide is comprised of sialic acid, fucose, hexosamines, or sulphated hexosamines; Vitamin A; retinolic acid derivatives; retinol derivatives; taurine; and glutamic acid and its conjugates. The composition may also additionally comprise at least one compound selected from the group consisting of ammonia; primary alkyl amines; secondary alkyl amines; tertiary alkyl amines; and a carboxylic acid R 6 CO 2 H, wherein R 6 is C 1 -C 30 alkyl that is saturated or unsaturated, and oxidized and/or hydroxylized derivatives thereof.

The method of the invention also embraces stimulation of TNF production by administration of a composition comprising at least one compound selected from the group consisting of taurocholic acid and its sulphated derivatives; glycocholic acid and its sulphated derivatives; sphingosine; a diacyl glycerol; lecithin; an oligosaccharide of less than 10 saccharide units in length, where said oligosaccharide is comprised of sialic acid, fucose, hexosamines, or sulphated hexosamines; vitamin A; retinoic acid derivatives; retinol derivatives; taurine; and-glutamic acid and its conjugates.

The present invention also provides a method of treating pancreatic cancer comprising administering to a patient suffering from said cancer a therapeutically effective amount of the compositions of the invention.

Also forming part of the present invention are compositions comprising (1) micelles of sphingosine or sphingosine complexed with a salt, or (2) micelles of retinolic acid or its derivaties, which have at least one of the following properties:

(a) is capable Of stimulating monocytes and macrophages in vitro to produce one or more cytokines;

(b) is capable of stimulating monocytes or macrophages to produce tumor necrosis factor in vitro or in vivo; or

(c) has an anti-proliferative effect in a malignant mouse hybridoma cell line.

The micelles may also comprise a diacyl glyceride or lecithin, and may further comprise a bile acid salt, and a source of ammonium or alkyl ammonium ions.

Finally, the present invention also contemplates compositions comprising (1) sphingosine, a bile acid salt, and a source of ammonium or alkyl ammonium ions, (2) a bile acid salt, sphingosine, a diacyl glycerol, a source of ammonium or alkyl ammonium ions, and a retinal derivative, (3) a diacyl glyceride, lecithin, and a bile acid salt, or (4) (a) a diacyl glyceride, (b) lecithin, and (c) a mucin hydrolysis product or a proteoglycan hydrolysis product, which has at least one of the following properties:

(a) is capable of stimulating monocytes and macrophages in vitro to produce one or more cytokines;

(b) is capable of stimulating monocytes or macrophages to produce tumor necrosis factor in vitro or in viva; or

(c) has an anti-proliferative effect in a malignant mouse hybridoma cell line.

The following non-limiting examples are illustrative of the present invention:

EXAMPLE 1

Preparation of the Composition of the Invention

Bovine bile is collected from healthy herds at least one and one half years old which have been slaughtered for food use at a licensed and inspected abattoir. The gall bladders are collected from the slaughtered animals which have been inspected and the gall bladders are separated from the livers and examined by a veterinarian to confirm that the gall bladders are free of parasites and evidence of infection, and thus are suitable for use as a source of bile.

Gall bladders which pass this inspection are wiped with a solution of 70% ethanol to sanitize the exterior and a syringe is inserted to remove the bile. The bile removed is visually examined by the veterinarian in the syringe to assure that it contains no blood or pus and is otherwise satisfactory. Bile found to be satisfactory is transferred into a graduated amber bottle containing ethyl alcohol. The bile is a greenish fluid substantially free of blood and pus. Bile is added to each bottle to a level marked on the bottle, twice the level of ethanol present to give a 50% bile/ethanol solution. The bile/ethanol solution is a greenish fluid substantially free of foreign material in an approximate 50%/50% bile/ethyl solution. It also shows positive for ethyl alcohol USP XXII Part B. These bottles are labelled with the date of collection which serves as the lot number. A minimum of fifty animals serve as the pool for each lot. Fragments of livers, spleen, and lymph nodes are also collected from the animals whose bile made up the pool and the fragments are examined for the presence of parasites or other indications of disease.

The bile/alcohol mixture is then centrifuged at 4200 RPM for at least 2½ hours at 20+/−2° C. The supernatant liquid is decanted and checked for pH and ethanol content. The decanted liquid is then subjected to an activated charcoal treatment. The treated bile/ethanol is then monitored for Optical Density (“O.D.”) and conductivity. O.D. levels or conductivity levels outside acceptable specifications will require that the bile ethanol solution be given additional treatment with activated carbon to achieve a reading within specification limits.

Following activated carbon treatment, the solution is filtered through a filter (for example using filters having a 2.5 μm retention), the alcohol is evaporated off (for example, by heating to less than 85° C.) and the solution is concentrated to approximately one eighth of the original bile/ethanol solution volume. The concentrated solution is cooled to 20-25° C. This solution is then mixed with ethyl ether and the ether phase is discarded. This step may be repeated once. The aqueous phase is heated to remove residual ether (for example by heating up to about 55° C. for about 10 hours) and further reduced in volume to one tenth of the original bile/ethanol volume by heating to about 80-85° C. The resultant composition is then tested for appearance, biological activity and ethanol and ether content. The composition is a clear yellowish solution essentially free of foreign matter, and it contains not more than 10 ppm ethanol and not more than 5 ppm ether. In the bioassay described in Example 4, the non-proliferative growth is 18+/−Units/ml.

Identity and purity are determined using reverse-phase high pressure liquid chromatography. Potency is assayed using the antiproliferative method as described in Example 4.

Initial batches of the composition of the invention were manufactured as a non-buffered liquid. Subsequent batches were manufactured as a buffered liquid, prepared by adjusting the pH of the composition to about 7.4+/−0.05, using hydrochloric acid (1%) solution and sodium hydroxide (1% solution), as well as using dibasic and monobasic sodium phosphate salts as buffers. Bioburden reduction is conducted in a steam autoclave at 104+/−2° C. for 60 minutes. The bulk solution is filled into 5 ml or 10 ml sterile bottles and capped. The filled and capped bottles are subjected to three sterilization cycles by autoclaving them at 104° C.+/−2° C. for 60 minutes followed by incubation at 35° C. for 23+/−1 hrs. Between each cycle (autoclave plus incubation samples) are taken and tested for bioburden. Following the last cycle, the bottles are visually inspected against a black and a white background to detect any particulates which may be present.

Following inspection, the lot is sampled and tested for conformance to specifications. Tests include identity, sterility, pyrogenicity, endotoxin, bioassay, HPLC and general safety (See Table 1).

TABLE 1
Results for Individual Batches
FINAL	BATCH #	BATCH #	BATCH #
PRODUCT TEST	BC0226	BC0227	BC0228
Biological Activity	17	14.0	22.5
(18.0 +/− 5 units/ml)
Identity/Purity	Pass	Pass	Pass
Agrees with reference
Safety (Passes test)	Pass	Pass	Pass
Pyrogeniticy (temp. increase
shall not exceed 0.4° C.)
Endotoxin ≦0.4 EU/ml	≦0.25	≦0.25	≦0.25
Sterility (no growth)	Pass	Pass	Pass
pH (7.40 +/− 0.05)	7.45	7.39	7.36
Appearance - Visual (clear,	Pass	Pass	Pass
light yellowish liquid with
little or no precipitate)
Appearance - O.D (passes test)	0.088	0.118	0.088
Osmolarity	540	603	445
IN-PROCESS TEST
Solids (18 +/− 3 mg/ml)	18	15	20
Amino Acids	790	742	878
(800 +/− 10% mg/ml)
Ethyl Alcohol	Pass	Pass	Pass
(not more than 10 ppm)
Ethyl Ether	Pass	Pass	Pass
(not more than 10 ppm)
Conductivity	25	22	29
(26 +/− 5 mMHO)

EXAMPLE 2

Physical Chemical and Biochemical Characteristics of the Composition of the Invention

A number of the physicochemical characteristics of the preparation (conductivity, osmolarity and total solids) are shown in Table 2. More particularly, test results for three manufactured batches of a composition prepared in accordance with Example 1 were carried out. The results shown in Table 2 demonstrate the sterility, potency, and reproducibility of the manufactured product. It is noted that the ethyl alcohol and ethyl ether are measured as in-process tests only. A summary of the method for determination of the biological activity is provided in Example 4.

TABLE 2
Product Specification
Test	Specification	Method
Biological Activity	18 +/− 5 units/ml	Biological Activity
Identity/Purity	Agrees with reference	HPLC
Safety	Passes test	General safety test
		(mice and guinea pigs)
		21 CFR part 610.11
Pyrogenicity	Temperature increase	Pyrogen test (rabbits)
	shall not	USP
	exceed 0.4° C.
Endotoxin	<2 EU/ml	Limulus
		Amoebocyte Lysate
		Test USP
Sterility	No growth	Sterility Test
		USP
pH	7.40 +/− .05	pH test USP
Appearance	Clear, light yellowish	Visual Inspection
	liquid with little
	or no precipitate
Solids	18 +/− 3 mg/ml	Lyophilization
Amino Acids	800 +/− 10% ug/ml	Trinitrobenzene-
		sulfonic
		Acid Method
Osmolarity		Freezing point
		depression
		USP
Ethyl Alcohol	Not more than	Direct Injection Gas
	100 ppm	Chromatography
Ethyl Ether	Not more than	Direct Injection Gas
	100 ppm	Chromatography
Conductivity	26 +/− 5 mMHO	Copenhagen Radiometer
		Model

Physical and chemical properties such as conductivity, osmolarity and total solids are consistent with a composition of over 99% salt. Less than 1% of the solids in the composition is organic material, there are no lipids, around half are carbohydrates and the rest are amino acids. Proteins and peptides are present. SDS Gel electrophoresis confirmed that there may be more peptides than proteins in the composition. High molecular weights are not detected. This is an important feature of a peptide drug because it is not expected to be immunogenic.

HPLC and bioassay test methods for the composition of the invention were developed using the nonbuffered product. These tests are used to characterize the product as the buffered liquid and the concentrated formula. The HPLC results described below indicate that the product is the same in all of the presentations. The bioassay shows that the activity of the concentrated composition is two and a half times greater than the original composition. Therefore, the product used in the studies has been demonstrated as being equivalent.

Reversed Phase (C 18 ) HPLC Analyses of the Composition of the Invention

The composition of the invention has a consistently reproducible pattern on reversed phase HPLC in which peaks are seen early in the exclusion fraction and at about 27 and 32 minutes. Before, in-between and after the tall peaks, there are smaller peaks that vary in intensity. The HPLC readings for three lots of the concentrated composition of the invention are shown in FIGS. 1 to 3 . RP-HPLC profiles for batches B0211 (FIG. 4 ), B0209 (FIG. 5 ), B29/3006 ( FIG. 6 ) and B15/1606 (FIG. 7 ), also show a very reproducible pattern.

The RP-HPLC to characterize the composition of the invention was carried out as follows. Bio-Rad Hi-Pore RP 318 guard column (C 18 ), 4.6×30 mm (Bio-Rad) and Bio-Rad Hi-Pore RP 318 (C 18 ) column, 4.6×250 mm was used. The samples were dialyzed in 0.1% trifluoroacetic acid (TFA Pierce) in H 2 O (Buffer A) and applied to the column. Buffer A was run for 10 minutes, then a linear gradient 0-80% of Buffer B (0.1% TFA in 100% acetonitrile) was run for 55 minutes. At the end of this period, 80% Buffer B was run for 5 minutes and 80-0% of Buffer B for 5 minutes. Flow rate was 1.0 ml/minute. Fractions from successive runs were collected and pooled and concentrated in a Speedvac (Model SVC 200H, Savant Instruments, Farmington, N.Y.).

Preliminary Characterization of the Composition

Peaks from HPLC were submitted for protein sequencing. The initial sample, the major HPLC peak designated RP-HPLC −31.00 min, batch 0210 in TFA/CH 3 CN, failed to yield data when subjected to N-terminal sequence analysis. The results may be interpreted as either a quantity problem or N-terminal blockage. Quantitation of protein content of that fraction by amino acid analysis after acid hydrolysis revealed that sufficient quantity should have been subjected to sequence analysis; however, because of the composition it was thought the sample may not be a protein and thus N-terminal blockage would not be the problem. The samples displayed the following composition: about 70% Glx (glutamate/glutamine) plus about 15% glycinen (Gly). Furthermore, analysis of the equivalent sample from RP-HPLC gave similar results: 68% Glx and 15% Gly (Table 3). Further characterizations of unfractionated material plus several other fractions revealed the following. The starting material (32 mg/ml) yielded a sequence signal indicating a polyglutamate peptide/protein, consistent with the amino acid composition data.

TABLE 3
Mole % Glx Gly - HPLC Fraction run R1 (see profile)
Fraction	Analysis #	Mole %	SUM
#	(07/11/92)	Glx	Gly	Glx + Gly
1a	393	28	44	72
1b + c	394	35	37	72
2a	395	39	36	75
2b	396	35	43	78
2c	397	44	30	74
3a	398	69	21	90
3b	399	70	10	80
3c	400	52	18	70
4	401	68	15	83
	402	48	30	78

Analysis of several HPLC fractions (Table 3) revealed that all are very rich in Glx (Glu/Gln) (28-70 mol %) and Gly (10-44 mol %). Each fraction, however, displayed real differences in the relative amounts of the other amino acids.

EXAMPLE 3

Biological Activity of Fractions of the Composition

The biological activity of fractions of the composition have been investigated. The biological activity of the composition is thought to be attributable to small molecular weight components (m.w. less than 3000 daltons). This was determined through an experiment in which four fractions of the composition and unfractionated composition were tested for biological activity. The first fraction contained proteins and peptides with molecular weight less than 3000 daltons, while the remaining three fractions contained additional larger molecular weight proteins and polypeptides. All fractions contained additional larger molecular weight proteins and polypeptides. All fractions and unfractionated compositions demonstrated the same biological activity. Since all fractions were as effective as the unfractionated product, and since the common denominator of all fractions was the presence of the same concentration of molecules smaller than 3000 daltons, this led to the conclusion that the biological activity of the composition is due to components with molecular weights smaller than 3000 Daltons.

EXAMPLE 4

Effect on Malignant Cell Lines

The effect of the composition of the invention on the proliferation of cultures of four malignant cell lines (Daudi-human lymphoma cells, ME-180 human cervical carcinoma cells, T-24-human bladder carcinoma cells, and mouse hybridoma cells #6-1) was measured. The composition of the invention had an antiproliferative effect on the mouse hybridoma cells. The studies suggested that the inhibitory effect of the composition in mouse hybridoma cells is antiproliferative rather than cytocidal.

A bioassay based on the reproducible antiproliferative effect of the composition of the invention was designed in a mouse hybridoma cell model to facilitate characterization of the composition. The bioassay is carried out as follows. The osmolarity and pH of the composition are adjusted to match that of the cell culture medium in order to isolate the composition's biological activity from its physical and chemical properties. Serial dilutions of isotonic composition from 1:5 to 1:10,000 are-prepared in culture medium. Hybridoma cell samples are specifically quantitated. Using a hemocytometer the cells in a 100 μl sample are counted. The cells are concentrated by centrifugation and then cell concentration is adjusted to 1,000 cells/ml (twice the final desired concentration) by addition of appropriate volumes of fresh media. Hybridoma cell suspensions (1 ml) mixed with corresponding dilutions of the composition (1 ml) are incubated in 24 well plates at 37° C. in a humid atmosphere with CO 2 controlled at 6%. After 96 hours, each well is sampled at 100 μl×3 and placed in a 96 well plate. (A blank of 100 μl of medium without cells is included.) Cell density is determined using a PROMEGA CellTiter 96™ kit. Cell concentration is measured by reading absorbance at 595 nm (650 nm reference) and recorded by ELISA plate reader. For each assay, a standard curve of cell concentration is prepared. Cultured cells in the log phase of growth are sampled, counted, concentrated and resuspended in serial dilutions. Each dilution is sampled at 100 μl×3 and placed in a 96 well plate. The standard curve is constructed by plotting cell density versus “net” OD at 595 nm after subtraction of zero cell blanks and OD at 650 nm. Cell density of unknown samples is determined by interpolation.

To calculate biological activity, cell density as a percentage of the control (no composition) is plotted against final composition dilution (in log and linear scale), and a curve is fitted through the points utilizing the Spline curve fitting method. Then, the composition dilution that corresponds to 50% inhibition of cell proliferation is determined manually and converted directly to UNITS of composition. By definition, one unit of composition inhibits by 50% the proliferation of 1 ml cell culture of cell line HYB #6-1, seeded at 500 cells/ml, after 96 hours at 37° C. and 6%CO 2 . The average activity of the composition of the invention in the bioassay is 18.24+/−1.82 Units/ml.

EXAMPLE 5

Effect on T and B Lymphocytes in Culture

The composition of the invention has been shown to be non-toxic to normal T and B lymphocytes in culture. The growth of human lymphocytes was examined under carefully controlled conditions in the presence and absence of the composition. Standard concentrations of lymphocytes were incubated in wells containing various concentrations of the composition. When normal T and B human lymphocytes were incubated with the composition in concentrations similar to those that are used clinically, there were no adverse effects as judged by Trypan Blue exclusion.

The effect of the composition on the survival of human peripheral blood mononuclear cells (PBMN) was examined. In this experiment, PBMN were incubated for 24 and 48 hrs in plastic microwell plates with various volumes of the composition and tissue culture medium. At the end of this period, the number of surviving cells was estimated by trypan blue dye exclusion. Table 4 shows that the number of surviving cells fell at 24 and again at 48 hours; however, the number of surviving cells in the presence or absence of the composition was not different. Moreover, increasing volumes of the composition had no effect on survival (Table 4). Thus, the composition showed no cytotoxicity to human PBMN.

TABLE 4
Concentration of Viable PBMN After Incubation
	No. of Live PBMN per Well by Trypan Blue (×10 6 ) 1
	Zero	After 24 hrs No.	After 48 hrs No.
	time	% viable	% viable
Patient S.Z.
Concentration
(μl/well)
0	0.70 2	0.23 (33)	0.10 (14)
25		0.43 (61)	0.15 (21)
50		0.10 (14)	0.23 (33)
100		0.15 (21)	0.18 (26)
200		0.48 (69)	0.23 (33)
LPS (μg/well)
1		0.30 (43)	0.28 (40)
10		0.25 (36)	0.13 (18)
Patient E.S.
Concentration
(μl/well)
0	1.30 2	0.70 (54)	0.33 (25)
25		0.65 (50)	0.15 (12)
50		0.68 (52)	0.38 (29)
100		0.75 (58)	0.23 (18)
200		0.65 (50)	0.20 (15)
LPS (μg/well)
1		0.60 (46)	0.53 (41)
10		0.15 (12)	0.15 (12)
1 Approximately 1 × 10 6 cells plated/well in triplicate.
2 Actual number of cells counted/well (×10 6 ).

EXAMPLE 6

Cytokine Content of Composition

ELISA assays for TNF-α, IL-1, IL-2, IL-4, IL-6, IL-8, GM-CSF and IFN were conducted on the composition of the present invention. It was determined that the composition of the invention contained no measurable levels of cytokines (TNF, IL-1 alpha, IL-1 beta, IL-4, IL-6, IL-8, GM-CSF and IFN gamma) (See Table 5).

TABLE 5
ELISA DETERMINATION OF
CYTOKINES IN COMPOSITION
	Cytokine	50 μl	100 μl
	TNF pg/ml	<5	<5
		Detection Limit:
		5 pg/ml
	IL-1β pg/ml	—	6.5
		Detection Limit:
		4.3 pg/ml
	GM-CSF pg/ml	<5	—
	IL-6 pg/ml	<7	—
		Detection Limit:
		7 pg/ml
	IFNγ pg/ml	<5	—
		Detection Limit:
		5 pg/ml
	IL-1α pg/ml	<50	—
		Detection Limit:
		50 pg/ml
	IL-4 pg/ml	—	<3
		Detection Limit:
		3 pg/ml
	IL-8 ng/ml		<4.7
		Detection Limit:
		4.7 ng/ml

EXAMPLE 7

Physical, chemical, and biological properties, were determined for a number of batches of the composition of the invention prepared in accordance with the method as described in Example 1. In addition, the chemical composition of the batches was determined and an amino acid analysis of the batches was conducted. The results are shown in Tables 6-8.

TABLE 6
CHEMICAL COMPOSITION
					High M.W.
	Solids	Amino Acids	Sugars	Lipids	>3 kD PROT
Batch No.	mg/ml	μg/ml	μg/ml	μg/ml	μg/ml
B0201	15.3	4.59	40.85	ND < 0.5	NA
B0202	15.7	13.16	54.95	ND < 0.5	NA
B0203	15.0	72.67	25.5	ND < 0.5	NA
B0208	7.8	4.53	30	ND < 0.5	ND < 1.0
B0209	8.5	2.27	24	ND < 0.5	ND < 1.0
B0211	5.6	1.47	19.2	ND < 0.5	ND < 1.0
B0106	32.2	1.16	32.6	ND < 0.5	ND < 1.0
B0706	32.7	I.42	26.2	ND < 0.5	ND < 1.0
B1306	22.3	8.01	48	ND < 0.5	ND < 1.0
B2006	21.7	9.73	38.4	ND < 0.5	ND < 1.0
B2306	28.5	16.35	42	ND < 0.5	ND < 1.0
PHYSICAL, CHEMICAL AND BIOLOGICAL PROPERTIES
				Absorban.	UV,
Batch		Conduct.	Osmolar.	O.D.	VIS	Activity
No.	pH	mMHO	mOsM	280 nm	Peaks	Units/ml
B0201	7.37	16.9	361	0.98	404 nm	10.5
B0202	7.35	17.3	298	0.777	None	6.5
B0203	7.3	17.7	360	0.67	365 nm	21.0
B0208	7.00	16.1	250	0.453	None	8.1
B0209	7.31	11.2	259	0.594	None	6.7
B0211	7.35	34.9	175	0.287	None	7.5
B0106	7.57	34.3	627	0.341	None	17.2
B0706	7.57	11.6	627	0.387	None	23.0
B1306	8.02	35.6	790	1.147	None	17.0
B2006	8.56	33.9	651	1.024	None	21.0
B2306	8.01	35.1	623	1.054	None	19.0
Comments:
1. To batches No. B0106 and B0706 full isotonic PBS solids were added.
2. Batches B1306, B2006 & B2306 were concentrated X2, no pH adjustment.

TABLE 7
CHEMICAL COMPOSITION
					High M.W.
	Solids	Amino Acids	Sugars	Lipids	>3.5 kD PROT
Batch No.	mg/ml	μg/ml	μg/ml	μg/ml	μg/ml
B0213	31.6	21	61	ND	ND
R0201/−pH	52.5	1553	216	ND	ND
R0201/+pH	55.8	1530	280	ND	ND
C0203	36.1	113	42	ND	ND
0-13/2109	12.1	149	36	ND	ND
B27/2806	17.5	28	37	ND	ND
B29/3006	28.7	26	60	ND	ND
B15/1606	26.8	41	45	75	ND
PHYSICAL, CHEMICAL AND BIOLOGICAL PROPERTIES
Batch		Conduct.	Osmolar,	Absorban.	UV, VIS	Activity
No.	pH	mMHO	mOsM	O.D. 280 nm	Peaks	Units/ml
B0213	7.75	29.5	628	0.48	none	14.5
R0201−pH	7.95	44.5	877	1.59	271 nm	51.5
					0.65 O.D.
R0201/+pH	7.60	50.0	1162	2.29	266 nm	61.5
					1.6 O.D.
C0203	7.90	34.8	657	0.96	NONE	5.0
0-13/2109	7.73	17.0	316	0.83	none	14.5
B27/2806	7.71	22.0	453	0.49	none	12.4
B29/3006	7.67	28.8	605	0.55	none	14.0
B15/1606	7.84	35.0	753	1.04	none	14.0

TABLE 8
AMINO-ACID COMPOSITION
Batch #	B-0208	B-0209	B-0211	01/06	07/06	1306	2006	2306
Asparagine	365				113			289
Serine		69	12	7	17	144	119	308
Glycine	22	449	274	279	417	3731	5314	10371
Histidine		192		90	68	938	1335	2114
Arginine				161	533
Threonine		19	13		30	148	142	250
Alanine		173	112	24	64	949	1002	1423
Proline	1092				74	817	639	1075
Tyrosine	15	55	57	43	39	205	135	45
Valine	121	63	31	10	15	367	335	224
Methionine		970	461	462	13	107	121	70
Cysteine		103	90	41	12	86	49	10
Isoleucine	2721	84	95	17		232	216	68
Leucine		58			9	221	242	84
Phenylalanine		57	200	16		45	80	23
Lysine	191	36	123	6	18	15
Total AA	4.53	2.27	1.47	1.16	1.42	8.01	9.73	16.35
μg/ml

EXAMPLE 8

Activation of Monocytes and Macrophages

Investigations have shown that the composition of the invention will activate normal monocytes to demonstrate cytotoxicity towards the Chang hepatoma cell line which is used to measure monocyte toxicity and that the monocytes and macrophages from cancer patients (cervical and ovarian cancer) have been stimulated by the composition to attack and destroy their own particular tumor cells.

More particularly, the monocyte tumoricidal function has been tested in the presence of the composition of the invention and the basic procedure for these experiments is outlined below.

Venous blood is collected in heparinized vacutainer tubes. The blood is diluted 3:1 in Hanks balanced salt solution (HBSS) layered onto lymphocyte separation medium and centrifuged to obtain a band of peripheral blood mononuclear cells (PBMN). After centrifugation, the mononuclear cell layer is recovered from the interface washed twice in medium and monocytes are enumerated by latex ingestion. Nonocytes are isolated by adherence in 96 well plates (for 2 hours at 37° C. followed by two cycles of washing). Adherent cells are estimated to be greater than 90% monocytes. Wells containing adherent cells are incubated overnight in the presence of the composition 11:10 dilution) granulocytemacrophage stimulation factor or PHA. Then adherent cells are washed and incubated overnight with tumor cells. For studies using a standard cell line 51 CR-labelled Chang hepatoma cells are used because this cell line is insensitive to natural killer cell cytotoxicity. These hepatoma target tumor cells are added to adherent cell monolayers at an effector:target (E:T) cell ratio of 20:1. This E:T ratio is used because it falls well into the plateau range on a curve prepared by varying the E:T ratio from 5:1 to 30:1.) After 24 hours supernatants are collected and 51 Cr release is quantitated. The percent specific cytotoxicity is calculated as:

% specific release=( E−S )/( T−S )×100

where E=CPM released from target cells in the presence of effector cells; S=CPM released from target cells in the absence of effector cells; T=CPM released from target cells after treatment with 2% sodium dodecyl sulfate.

Using this protocol, the composition was found to cause monocytes from healthy donors to exert cytotoxicity toward the Chang hepatoma cell line. Subsequently, whether monocytes and macrophages from a cancer patient could be stimulated by the composition to attack and destroy their own particular tumor was investigated. Using similar protocols as described for the standard cell line (Chang hepatoma cells), monocytes and/or peritoneal macrophages from cancer patients were isolated. (Peritoneal macrophages were isolated from peritoneal fluids collected at the time of laparoscopy). The composition was found to activate peripheral monocytes and peritoneal macrophages from a patient with cervical cancer to produce cytotoxicity against the patient's own tumor cells. This effect was comparable to or better than that produced by interferon or lipopolysaccharide. Peritoneal macrophages from a patient with ovarian cancer were also found to be stimulated by the composition to attack and destroy the ovarian tumor cells in culture.

EXAMPLE 9

Effect on TNF

Studies were conducted to evaluate the effect of the composition of the invention on cytokine release from peripheral blood mononuclear cells (PBMN). ELISA assays for TNF-α, IL-1a, IL-2, IL-4, IL-6, IL-8, GM-CSF and IFN were conducted.

The following methods were used in the studies described in the Example.

In the studies of TNF, whole blood was drawn from 5 healthy subjects into heparinized Vacutainer tubes. Peripheral blood mononuclear cells (PBMNs) were isolated by gradient centrifugation on Ficoll-Hypaque (Pharmacia). The PBMN were washed twice with phosphate buffered saline (PBS), counted and resuspended in RPMI 1640 culture medium (Gibco Labs) at a concentration of 10 6 cells/0.5 ml. These cells were cultured in 24 well, flat-bottomed tissue culture plates (Falcon, Becton, Dickinson). Of the PBMN suspension, 0.5 ml was added to each well containing 50 ng Lipopolysaccharide (LPS) (from E. coli ), 10 μl fetal calf serum and the respective volumes of composition tested 10-300 μl). To neutralize the hyperosmolar effect of the composition, distilled water was added to the culture wells at a volume equivalent to 10% of the volume of composition used. The total volume was then made up to 1 ml/well with RPMI. As control, PBS was used instead of composition. The cells were cultured for 2, 6, 24, 48 and 72 hours at 37° C. in a humidified 5% CO 2 incubator. At the end of each incubation period, the cells were harvested and cell free culture fluids were obtained by centrifugation at 9000 rpm for 10 minutes. The samples were then stored at −70° C. until ELISA for cytokines was carried out (within 2 weeks).

Protein estimation of the composition was done using the Pierce Micro BCA Protein determination technique (Smith et al., Anal. Biochem. 1985, 150:76-85). 10 μl of a sample of the composition was made up to 1 ml with distilled water. Five concentrations of Bovine Serum Albumin (0.150 μg/ml) was also made up to be used as standards. As a blank, 0.1N NaOH was used. To all these samples was added a mixture of BCA, 2% Bicinchonic Acid sodium salt) (Pierce), Copper Sulfate 4% and Microreagent A (NaCO 3 , NaHCO 3 , Na tartrate in 0.2N NaOH). The sample mixtures were incubated for 1 hr at 60° C., cooled and the resultant absorbency read at 562 nm using a spectrophotometer. The amount of protein in the test sample was then compared to the plotted standard curve and the appropriate calculations made. The protein concentration of the composition was found to be low and estimated to be 32 μg/ml.

Cytokine synthesis in the supernatants were measured after stimulating human PBMN with the composition of the invention at volumes of 200 and 300 μl/well. The initial preparations of the composition show no stimulatory effect on cytokine production (Table 9). If there was any effect there was the suggestion that cytokine production was below the constitutive level when PBMN were incubated in medium alone.

TABLE 9
Direct Effect of Composition on Cytokine Production after 24 hrs
	Amount of Cytokine Released (pg/ml) 1
Cytokine		Composition	LPS
Assayed	Medium	100 μl	200 μl	1 μg
IL-1α	61.6 ± 12	59.6 ± 7.8	54.3 ± 6.0	315 ± 117
IL-1β	199 ± 184	218 ± 165	188 ± 174	965 ± 99
TNF 2	203 ± 149	151 ± 117	107 ± 120	1501 ± 284
IL-6	928 ± 776	853 ± 673	829 ± 543	2016 ± 41
IL-8	126 ± 70 3	94 ± 50 3	77 ± 41 3	361 ± 165 3
GM-CSF	13 ± 4	13 ± 7	15 ± 11	54 ± 20
IFN-γ	11 ± 18	9 ± 14	5 ± 6	54 ± 94
IL-4	<3.0	<3.0	<3.0	<3.0
1 Mean of eight patient samples in duplicate
2 Mean of seven patient samples in duplicate
3 ng/ml

Experiments were performed to determine whether the composition of the invention would impair LPS-stimulated release. LPS was used as a positive stimulus, and the ability of the composition to impair LPS stimulated release of cytokines was compared for the different cytokines (Table 10). The composition clearly inhibits IL-1 alpha, IL-1 beta and TNF. The effects on the other cytokines IL-6, IL-8, IFN-gamma and GM-CSF were not as marked. However, in no instance did the compositions tested augment the effect of LPS stimulated release of cytokines.

TABLE 10
Difference between LPS Released Cytokine
and LPS plus Composition
Stimulant of Cytokine
	Cytokine	LPS	LPS +	Mean
	Assayed 1	(50 ng/ml	Composition 2	Difference
	IL-1α	129 ± 135	77 ± 27	−52
	IL-16	1314 ± 723	919 ± 460	−395
	TNF	915 ± 763	497 ± 525	−418
	IL-6	2320 ± 1081	2320 ± 1145	0
	IL-8	118 ± 62	109 ± 56	−9
	(ng/ml)
	IFN-γ	30 ± 24	13 ± 10	−17
	GM-CSF	54 ± 65	50 ± 63	−4
	1 PBMN from six patients were tested in duplicate
	2 Composition batch B0209

The effect of different volumes of the compositions of the invention in inhibiting LPS stimulated release of TNF was examined. FIG. 8 shows a dose-response curve of the composition inhibiting release of TNF by PBMN stimulated with LPS. Ten μl of the composition inhibited about 10% and increased close to 30% inhibition at 100 μl of the composition. Another batch (B0201) of the composition inhibited LPS-induced TNF production at 200, 100 and 10 μl by 45, 21 and 12 percent, respectively.

Similarly, IL-1 beta production was inhibited in a dose-dependent manner by the composition: 100, 25 and 10 μl of the composition inhibited by 16, 10 and 9 percent, respectively.

Different batches of the composition were examined for their effect on LPS-induced release of TNF. In summary, it was found that batches of the composition produced in the same way and from the same animal induced an identical effect. However, changes in the method of preparation or the composition from different animal species had different effects. Batches B29/3006, B0213 (B=bovine) and C0203 (goat) induced a strong release of TNF above that induced by LPS alone (Table 11).

TABLE 11
Dose-Response Effect on TNF Release of Strong
Stimulatory Batches of Composition
	Difference in TNFα Release Between
	LPS + Composition − LPS Alone
	Composition
Batch	Volume (μl)	TNF (pg/ml)
B0213	10	193 ± 161
	100	858 ± 819
	200	2131 ± 1742
B29/3006	10	121 ± 102
	50	422 ± 78
	100	834 ± 811
	200	2252 ± 676
C0203	10	101 ± 47
	50	643 ± 231
	100	2650 ± 1372
	200	1851 ± 980

Table 12 shows the results with batches B15/1606 (concentrated preparation) and B27/2806, which were moderately stimulatory.

TABLE 12
Dose Response Effect on TNF Release of Moderate Stimulatory
Batches
	Difference in TNFα Release Between
	LPS + Composition − LPS Alone
	Composition
Batch	Volume (μl)	TNF (pg/ml)
B27/2806	10	−24 ± 120
	50	71 ± 103
	100	667 ± 844
	200	984 ± 200
B15/1606	10	299 ± 351
	50	294 ± 145
	100	667 ± 800
	200	1224 ± 446

Table 13 shows that batch 013/2109 (sheep) was minimally stimulatory.

TABLE 13
Dose-Response Effect on TNF Release of Minimal Stimulatory Batch
	Difference in TNFα Release Between
	LPS +   − LPS Alone
Batch	Volume (μl)	TNF (pg/ml)
013/2109	50	−9 ± 73
	200	179 ± 162
	300	178 ± 373

Table 14 shows that batch R0201 (shark) was inhibitory at most concentrations for LPS-induced TNF production.

TABLE 14
Dose-Response Effect on TNF Release of Inhibitory Batch
	Difference in TNFα Release Between
	LPS +   − LPS Alone
Batch	Volume (μl)	TNF (pg/ml)
R0201	50	145 ± 256
	200	−370 ± 385
	300	−400 ± 185

Initially, the composition was shown to affect LPS-induced release of TNF from human PBMN. Thus, in the next series of experiments, the time effect of the composition on LPS-induced release of TNF was examined (Table 15). LPS stimulated release of TNF. By 2 hours the level had risen to 697 pg/ml and peaked at 6 hrs at about 2006 pg/ml. At 24, 48 and 72 hrs the release of TNF progressively fell. In fact, by 48 and 72 hrs, the TNF release was just above constitutive production levels. By contrast, Batch 0213 of the composition, which was strongly stimulatory for TNF release, showed no release of TNF above that produced by LPS alone at 2 and 6 hrs. Whereas LPS induced peak release of TNF at 6 hrs, the composition in combination with LPS induced peak release at 24 hrs at a time when the stimulatory effect of LPS had begun to fall. Unlike LPS alone, composition+LPS continued to stimulate TNF release at 48 and 72 hrs although the quantity of TNF released fell progressively (Table 15). Thus batch 0213 was stimulatory for TNF release. Batch B15/1606, which was only moderately stimulatory, inhibited LPS-induced release at 2 and 6 hrs. At 24 hrs, B15/1606 in combination with LPS was mildly stimulatory for TNF-release. At 48 and 72 hrs, B15/1606 in combination with LPS, had a mild stimulatory effect on TNF release. Thus, batch B15/1606 had a biphasic effect; early it inhibited LPS induced TNF release, and mainly at 24 hrs combination with LPS, had a mild stimulatory effect on TNF release. Thus, batch B15/1606 had a biphasic effect; early it inhibited LPS induced TNF release, and mainly at 24 hrs it caused a mild additive effect in conjunction with LPS in inducing TNF-release.

TABLE 15
Time Effect on Different Batches on TNF Release by LPS
		Mean Difference in TNFα (pg/ml)
		Release Between LPS + Composition
Time	TNF Released by	LPS Alone by Three Different Batches
(hr)	LPS (50 ng/ml)	BO213	B15/1606	R0201
2	697 ± 94	693 ± 339	363 ± 189	62 ± 42
6	2006 ± 736	1949 ± 422	1080 ± 377	430 ± 260
24	800 ± 222	2301 ± 658	876 ± 351	343 ± 183
48	170 ± 149	1419 ± 447	234 ± 183	129 ± 78
72	132 ± 147	945 ± 367	184 ± 107	153 ± 68

Batch R0201, which was inhibitory for LPS-induced release of TNF, was markedly inhibitory for LPS induced TNF release at 2, 6 and 24 hrs. At 48 and 72 hrs, LPS induced minimal TNF release and batch R0201 had minimal positive or negative affects at these times.

The direct stimulatory effect of Batch B0203 was tested at different volumes and then at different times.

Batch B0203 stimulated maximum TNF-release at about 100 μl (Table 16). The maximum effect was observed at 24 hrs for a volume of 200 μl (Table 17). Thus, the composition was able to stimulate TNF-release on its own, that is in the absence of LPS and the curves for TNF-release were similar to when the composition and LPS were combined. It appears that the composition by itself was less stipulatory for TNF release than the additive effect it had when combined with LPS.

TABLE 16
Dose-Response of Batch B0203 on Stimulating Release of TNF
Volume  Mean Amount of
(pg/ml) TNF Released at
(μl)    24 hrs
None    128 ± 207
(PBS 50 μl)
50      223 ± 65
100     327 ± 90
200     105 ± 54
300     189 ± 94

TABLE 16
Dose-Response of Batch B0203 on Stimulating Release of TNF
Volume  Mean Amount of
(pg/ml) TNF Released at
(μl)    24 hrs
None    128 ± 207
(PBS 50 μl)
50      223 ± 65
100     327 ± 90
200     105 ± 54
300     189 ± 94

In summary, the experimental results indicate that some batches of the composition are able to inhibit TNF release when human PBMN-are stimulated with LPS. Other batches have biphasic effects suggesting that they partially inhibit LPS-induced TNF release and have, as a late effect, the ability to induce a mild release of TNF. A third preparation had no inhibitory effect on LPS-induced release of TNF; but at a different time point than LPS, the preparation was able to stimulate human PBMN to release TNF. In conclusion, the composition of the invention can modulate TNF production, an important mediator of antitumor responses. A summary of the data is shown in FIG. 9 and in Table 18.

TABLE 18
TNF Bioassay Results
Batch	RP-HPLC		Normal or		TNF
No.	Peaks 1 Min	Source	Concentrated	Buffer	Release
B0213	27, 32	Bovine	Normal	Yes	↑ ↑
C0203	27, 32	Caprine	Normal	Yes	↑ ↑
013/2109	27, 32, 21:50,	Ovine	Concentrated	Yes	↓ / ↑
	25
R0201	21-25, 28, 29,	Shark	Normal	Yes	↓ ↓
	29.5, 27, 32
B29/3006	27, 32	Bovine	Normal	Yes	↑ ↑
B27/2806	↓27, ↓32	Bovine	Normal	Yes	↑
B15/1606	27, 32, 22, 28	Bovine	Concentrated	Yes	↑
				Yes	↓

EXAMPLE 10

This Example demonstrates, in summary, the following: The composition has TNF-α releasing activity and the TNF-α releasing activity is not related to any contamination with endotoxin. Priming of macrophages enhances the ability of the composition to stimulate release of TNF-α. The hyperosmolarity of the composition is not responsible for TNF-α releasing activity. The TNF-α releasing activity of the composition can be separated, in part, from other constituents. The TNF-α releasing activity of the composition does not bind or binds poorly to C 18 RP-HPLC. Most of compositions activity elutes early from RP-HPLC. Less than 20% of the activity of the composition is recoverable from the fractions that are retained on the RP-HPLC and elute later. The TNF-α releasing activity is precipitated by 80% acetonitrile, a high content of organic buffer. The precipitated material when reconstituted in aqueous buffer and analyzed on RP-HPLC shows great similarity to the excluded peak on RP-HPLC of the composition. It is possible to separate and concentrate the active component of the composition in a fraction that constitutes about 30% of the original material.

A. Polymyzin and TNF-alpha Release

To eliminate any possibility of an endotoxin effect of the composition experiments were performed with Polymyxin added to the reactants. Polymyxin inhibits the action of endotoxin on leukocytes. Table 19 shows that polymyxin completely inhibits the LPS-induced release of TNF-α. In the absence of polymyxin, LPS induces 517 pg/ml of TNF-α, whereas in the presence of Polymyxin 11 pg/ml of TNF-α is released. The composition, on the other hand, releases 1591 pg/ml of TNF-α in the presence of Polymyxin. In the absence of Polymyxin, LPS and the composition show more than just an additive effect of the stimulators, suggesting that the composition acts with greater intensity when macrophages are primed.

TABLE 19
Effect of Polymyxin on TNF Release by LPS + Composition
	Sample		TNF Released (pg/ml)
	Tested	Additive	Total	−LPS
	LPS	Polymyxin	11 ± 7	0
		None	517 ± 118	0
	Composition	Polymyxin	1591 ± 413	1581
	(#B0213)	None	5256 ± 2585	4738
	Notes:
	1. Total TNF Released is corrected for TNF release by 1640 Medium.
	2. Polymyxin concentration: 50,000 units/ml.
	3. Composition volume: 200 μl.
	4. With polymyxin, 8 patients tested. With no additive, 3 patients tested.
	5. LPS concentration: 50 ng/10 μl.

B. TNF-Releasing Activity of Reversed-Phased High Pressure Licruid Chromatoaraphy (RP-HPLC) Fractions

FIG. 10 shows the C 18 RP-HPLC profile of the comnposition, Batch 0213 and the 5 separate fractions that were tested for TNF-releasing activity.

Initially, the effect of different fractions were examined in the presence of Polymyxin. Table 20 shows that the early eluting fraction (2:05 to 21:20 minutes) from RP-HPLC had most of the detectable TNF-releasing activity. However, this activity was only about 50% of the starting composition. The right hand column of Table 20 shows the osmolarity of the samples. Batch B0213 was 369 and high. Fractions from 21 minutes and later had normal osmolarity, whereas the first fraction which was active was even more hyperosmolar than the starting material, indicating that much of the salt in the composition also eluted early. Therefore, whether the hyperosmolarity of the samples was inhibiting or enhancing TNF-α release was investigated as set out below.

TABLE 20
Separation and Testing of Different HPLC Fractions
of Composition in the Presence of Polymyxin
Sample	TNF Released (pg/ml)	Osmolarity
Tested	Total	−LPS	(mOsm)
1640 RPMI Medium	0	—	287
LPS (50 ng/10 μl)	40 ± 25	0
Composition #B0213	1222 ± 448	1182	369
Fractions (minutes)
2:05-21:20	554 ± 394	514	434
21:20-25:32	7 ± 7	0	301
25:32-31:55	0	0	299
31:55-34:58	4 ± 4	0	292
34:58-46:55	32 ± 28	0	295
Note:
1. Total TNF Released is corrected for TNF release by RPMI Medium.
2. Fractions are reconstituted in PBS.
3. Volume of composition and fractions: 200 μl.
4. Polymyxin concentration: 50,000 units/ml.
5. Number of patients tested: 5.
6. Osmolarity not measured for LPS.

Table 21 shows additional testing on two patients with composition fractions 1 (2:05-21:20) and 2 (21:20-25.32). Once more, most of the activity was recovered in fraction 1, but there was some activity in fraction 2. However, fraction 1 and 2 had only about 50% of the starting activity of the composition.

TABLE 21
Repeat Evaluation of Composition HPLC Fractions
2:05-21:20 and 21:20-25:32 minutes for release
of TNF in the Presence of Polymyxin.
	Sample	TNF Released (pg/ml)
	Tested	Total	−LPS
	1640 RPMI Medium	0	—
	LPS (50 ng/10 μl)	31 ± 31	0
	Composition #B0213	2013 ± 726	1983
	Fractions (minutes):
	2:05-21:20	526 ± 126	496
	21:20-25:32	132 ± 108	101
	Note:
	1. Total TNF Released is corrected for release by RPMI Medium.
	2. Fractions are reconstituted in PBS.
	3. Volume of composition and fractions: 200 μl.
	4. Polymyxin concentration: 50.000 units/ml.
	5. Number of patients tested: 2

To determine whether there was any nonspecific activity in the fractions, a blank run, was made and the same fractions were collected, concentrated and tested. The blank fractions produced virtually no TNF release. This result indicated that fractions from RP-HPLC could be used to test for TNF-α releasing activity without concern that the column or buffers contributed to TNF-α releasing activity.

Next the fractions of the composition were tested in the absence of Polymyxin in order to have the priming effect of LPS. Table 22 shows that Batch B0213 induced a marked release of TNF-α. There were 5 fractions of the composition from the RP-HPLC, with the elution times as indicated. Once again, fraction 1 had the most TNF-α releasing activity. However, with the priming effect of LPS, fractions 2 through 4 had some TNF-α releasing activity. Fraction 2 (21:20-25:32) had about 25% of the activity of fraction 1, and double the activity of the later fractions.

TABLE 22
Separation and Testing of HPLC Fractions of
Composition in the Absence of Polymyxin.
Sample	TNF Released (pg/ml)	Osmolarity
Tested	Total	−LPS	(mOsm)
1640 RPMI Medium	0	—	299
LPS (50 ng/10 μl)	219	0	305
Composition #B0213	1575 ± 470	1356	376
Fractions (minutes):
2:05-21:20	656 ± 206	436	321
21:20-25:32	345 ± 82	126	309
25:32-31:55	287 ± 70	68	305
31:55-34:58	262 ± 50	43	304
34:58-46:55	237 ± 59	18	376
1. Total TNF Released is corrected for release by RPMI Medium.
2. Fraction Reconstitution: 2:05-21:20 in water, 21:20 to 48:55 in PBS.
3. Volume of composition and fractions: 200 μl.
4. Number of patients tested for TNF release: 5.
5. Osmolarities are averages for 2 of 5 patients; standard errors very small, therefore, not reported.

Whereas Table 22 shows the results of using 200 μl volumes, Table 23 shows the results of testing an additional three patients with 100 μl of the composition and 100 μl of its RP-HPLC fractions in the absence of Polymyxin. Although the release by the composition is lower than with 200 μl of the composition (Table 22), the results are similar. Fraction 1 contains most of the activity with some activity in the later fractions 2 and 3.

TABLE 23
Separation and Testing of HPLC Fractions of
Composition in the Absence of Polymyxin.
Sample	TNF Released (pg/ml)	Osmolarity
Tested	Total	−LPS	(mOsm)
1640 RPMI Medium	0	—	303
LPS (50 ng/10 μl)	195 ± 72	0	302
Composition #B0213	692 ± 266	497	347
Fractions (minutes):
2:05-21:20	575 ± 82	379	310
21:20-25:32	226 ± 65	31	337
25:32-31:55	210 ± 71	14	305
31:55-34:58	192 ± 40	0	313
34:58-46:55	182 ± 73	0	344
1. Total TNF Released is corrected for release by RPMI Medium.
2. Fraction Reconstitution: 2:05-21:20 in water, 21:20 to 48:55 in PBS.
3. Volume of composition and fractions: 200 μl.
4. Number of patients tested for TNF release: 3.
5. Osmolarities are averages of patients tested; standard errors very small, therefore, not reported.

C. Effect of Osmolarity of TNF Release by the Composition

The composition of the invention is hyperosmolar. The effect of the hyperosmolarity of the composition on TNF-α releasing activity was studied. It was found that the composition, when adjusted for osmolality even to the point of being hypoosmolar, continued to release TNF-α.

D. Physicochemical Separation of the Composition by Precipitation With High Content of Organic Solvent

Since most of the TNF-releasing activity of the composition did not bind to the RP-HPLC as evidenced by its quick elution, it was decided to use a column that acts on the inverse principle of reversed-phase chromatography separation where the sample is in a high content of organic solvent and permits hydrophilic interaction. This separation technique is used for small polar substances. However, when the composition was brought to 80% acetonitrile a precipitate formed. Thus, some of the contents of the composition in a high organic solvent buffer precipitated. The precipitate and the soluble fraction were separated. Both the precipitate and soluble fraction were taken to dryness by lyophilization. The precipitate and soluble fraction were reconstituted in aqueous solutions and both analyzed by RP-HPLC and tested for TNF-α releasing activity. Table 24 shows that most of the TNF-α releasing activity was contained in the precipitated material.

TABLE 24
TNF-Releasing Activity of Fractions of Composition
Prepared by Precipitation in 80% Acetonitrile.
Sample	Volume/	TNF Released (pg/ml)	Osmolarity
Tested	Concentration	Total	−LPS	(mOsm)
1X 199 Medium			0	—	306
LPS	50	ng/	161 ± 50	0	301
	10	μl
Composition	100	μl	471 ± 304	310	307
#B0213	200	μl	505 ± 210	344	318
Supernatant	100	μl	192 ± 63	31	309
	200	μl	221 ± 69	60	310
Precipitate	100	μl	626 ± 212	465	307
	200	μl	1299 ± 565	1138	346
Note:
1. Total TNF Released is corrected for 1X 199 Medium.
2. Reconstitution: supernatant in PBS, precipitate in double distilled water.
3. Precipitate is in (hypotonic) 70% 1X 199 Medium.
4. Number of patients tested: 5.
5. Osmolarities are averages of 4 of 5 patients tested; standard errors very small, therefore, not reported.

RP-HPLc analy sis of both the precipitated ( FIG. 11 ) and soluble fractions ( FIG. 12 ) of the composition shows that the precipitate is principally the material contained in fraction 1 of the RP-HPLC of the composition (FIG. 10 ), and the soluble material contains the other fractions of the RP-HPLC of the composition (FIG. 10 ). Thus in two different ways, it was shown that the composition's activity is contained in the fraction that is minimally retained by RP-HPLC. In fact, the precipitate had equal activity to unprecipitated composition when tested at 100 and 200 μl.

Only the 200 μl precipitate had above normal osmolarity. Consequently, the precipitate and soluble fractions (supernatant) of the composition were separated by RP-HPLC ( FIGS. 11 and 12 ) and divided into two fractions (see profiles of RP-HPLC). Fraction 1 (2:00-21:10 min) was equivalent to the same fraction 1 of the composition separated by RP-HPLC and fraction 2 was equivalent to fractions 2 through 5 of the composition separated by RP-HPLC. The isolates were then tested for their TNF-α releasing activity. Table 26 shows that for two patients neither the precipitate or supernatant after RP-HPLC separation had any TNF-releasing activity. However, the results for the initial two patients (Table 25) suggested the possibility that the precipitate may not have been tested at the ideal volume. Consequently, the fractions were retested on two additional patients and at one lower concentration. Table 26 shows that at 50 and 100 μl, fraction 1 of the precipitate had the most activity. At 50 μl it released twice as much TNF-α as did 50 ng LPS. However, minor releasing activity was also found in RP-HPLC fraction 2 of the precipitate as well as minor activity was found on fractions 1 and 2 of RP-HPLC of the supernatant.

TABLE 25
TNF-Releasing Activity of HPLC Separated Fractions
of 80% Acetonitrile Precipitated Composition.
				Osmol-
Sample	199 Med-	Volume/	TNF Released (pg/ml)	arity
Tested	ium Used	Conc.	Total	−LPS	(mOsm)
Medium	1X			0	—	294
LPS	1X	50	ng/	194 ± 93	0	294
		10	μl
(#B0213)	1X-70% 1X	100	μl	855 ± 88	661	281
		200	μl	926 ± 163	732	291
Supernatant
2:15-21:28	1X	100	μl	92 ± 75	0	269
	1X	200	μl	25 ± 25	0	242
21:48-46:13	70% 1X	100	μl	61 ± 56	0	296
	70% 1X	200	μl	2 ± 2	0	297
Precipitate:
2:00-21:10	1X	100	μl	183 ± 15	0	305
	1X	200	μl	0	0	354
21:10-46:20	70% 1X	100	μl	62 ± 61	0	299
	70% 1X	200	μl	0	0	302
Note:
1. Total TNF released is corrected for TNF release by 199 Medium 1X.
2. Reconstitution: first fractions in double distilled water, second fractions in PBS.
3. Number of patients tested: 2.
4. Osmolarities are averages of patients tested: standard errors very small, therefore not reported.
5. Averages of values in 1X and 70% 1X 199 Medium.

TABLE 26
TNF Release with further Titraton of Precipitate and
Supernatant from 80% Acetonitrile Fractioned.
Sample	Volume/	TNF Released (pg/ml)	Osmolarity
Tested	Conc.	Total	−LPS	(mOsm)
1X 199 Medium			0	—	317
LPS	50	ng/	136 ± 38	0	320
	10	μl
#B0213	100	μl	274 ± 80	138	317
Supernatant
2:15-21:28	50	μl	171 ± 66	35	319
	100	μl	193 ± 73	57	322
21:48-46:13	50	μl	184 ± 34	48	311
	100	μl	162 ± 40	26	310
Precipitate
2:00-21:10	50	μl	287 ± 69	150	333
	100	μl	204 ± 40	68	355
21:10-46:20	50	μl	148 ± 46	11	323
	100	μl	198 ± 44	62	325
	200	μl	1299 ± 565	1138	346
Note:
1. Total TNF released is corrected for TNF release by 199 Medium 1X.
2. Reconstitution: first fractions in double distilled water, second fractions in PBS.
3. Number of patients tested: 2.
4. Osmolarities are averages of patients tested: standard errors very small, therefore, not reported.
5. All samples in 1X 199 Medium.

E. TNF-α Releasing by Different Media

It was observed that the seitch from RPMI 1640 to Medium 199 resulted in a lower TNF-α release was evaluated in Medium 199 and RPMI 1640 (Table 27). The results show that LPD from 10 to 200 ng is much more effective in releasing TNF-α in RPMI 1640 Medium than in Medium 199. Presumably the composition also gives greater release in RPMI 1640 Medium. Thus, cultural conditions can influence the degree of TNF-α release.

TABLE 27
Evaluation of TNF Release in Different Media
	1X 199 Medium	1640 RPMI Medium
	TNF		TNF
Sample	Released	Osmolarity	Released	Osmolarity
Tested	(pg/ml)	(mOsm)	(pg/ml)	(mOsm)
Medium	23	296	47
LPS:
10 ng/10 μl	124	291	—
50 ng/10 μl	155	292	356	285
100 ng/10 μl	147	293	323	289
200 ng/10 μl	213	294	455	288
1000 ng/10 μl	404	298	558	292
Note:
1. Number of patients tested: 1.

F. Osmolarity of the Composition

Table 28 shows the osmolarities of different batches of the composition. B0213 is moderately high at 675 mOsm. B0222 shown to have TNF-releasing activity even better than B0213 is less hyperosmolar, 581 mOsm. The fractions B0226, BC11-06 and BC11-09 range from 540 to 603 mOsm.

TABLE 28
OSMOLARITIES OF WHOLE BATCHES
			Osmolarity
	Batch #	pH	(mOsm)
	Concentrated:
	B0222	pre-pH	411
	B0222	pH adjusted	581
	B0216	pH adjusted	872
	B0219	pH adjusted	886
	Nonconcentrated:
	B0221	pre-pH	652
	B0221	pH adjusted	533
	B0213	pH adjusted	675
	B0225	pH adjusted	590
	B0226	pH adjusted	540
	BC 11-06	pH adjusted	445
	BC 11-09	pH adjusted	603

EXAMPLE 11

A. Tumor Necrosis Factor (TNF) Releasing Activity of a Composition of the Invention

1. Acetonitrile Precipitate and Supernatant of the Composition

As, shown in the prior Example, 80% acetonitrile precipitated material form a composition of the invention. The precipitated material and the unprecipitated (hence forth called Supernatant) composition were tested further to determine where the TNF-releasing activity resided.

Table 29 shows that the TNF-releasing component of the composition is precipitated by 80% acetonitrile, an organic solvent. Whereas the whole composition at 0.04 ml released about 15 pg/ml of TNF, 0.05 ml of precipitated Batches of the composition released 58 (B0222), O (B0221), and 17 (B0213) pg/ml, suggesting recovery of the TNF-releasing component by 80% acetonitrile precipitation. The precipitated composition was reconstituted in the same volume of liquid from which it had been precipitated. Thus, 0.1 ml of precipitate comes from 0.1 ml of whole composition and equals 0.1 ml whole composition.

TABLE 29
TNF Releasing Activity of Precipitates of Virulizin*
Prepared in 80% Acetonitrile.
	199	Quantity		Osmol-
Sample	Medium	in	TNF Released (pg/ml)	arity
Tested	Used	Wells	Total	−LPS	(mOsm)
LPS	1X	50 ng	322 ± 115	0	308
Virulizin: B0222
Whole	80% 1X	40 μl	337 ± 107	15	312
Precipitate	70% 1X	200 μl	1091 ± 137	769	351
	70% 1X	100 μl	620 ± 186	298	317
	70% 1X	50 μl	380 ± 132	58	297
	70% 1X	25 μl	312 ± 137	0	294
Virulizin: B0221
Whole	90% 1X	40 μl	282 ± 75	0	306
Precipitate	70% 1X	200 μl	981 ± 205	660	348
	70% 1X	100 μl	526 ± 169	205	314
	70% 1X	50 μl	308 ± 104	0	298
	70% 1X	25 μl	318 ± 185	0	292
Virulizin: B0213
Whole	90% 1X	40 μl	383 ± 72	61	312
Precipitate	70% 1X	200 μl	1143 ± 172	821	366
	70% 1X	100 μl	687 ± 186	365	326
	70% 1X	50 μl	339 ± 133	17	305
	70% 1X	25 μl	300 ± 144	0	298
Note:
1. Number of patients tested: 5.
2. Total TNF released is corrected for TNF release by 1X 199 Medium.
3. Osmolarities not corrected for 1X 199 Medium (311 mOsm).
4. Average of osmolarities given; standard errors not reported as values are very low.
5. Precipitate reconstituted in double distilled water.
6. LPS volume added to wells was 10 μl.
7. Wells contained a total volume of 1000 μl.
8. Sample volumes are equivalent.
*Composition of the invention is also referred to herein as VIRULIZIN

It is of interest to note that in earlier studies generally between 0.1 and 0.2 ml of the composition was used to stimulate TNF-release. The precipitates of B0222, B0221 and B0213, reconstituted to 0.1 and 0.2 ml released between 205 and 821 pg/ml of TNF. At 0.1 ml of precipitate, batches B0222, B0221 and B0213 released between 205 and 365 pg/ml of TNF, a very similar quantity of TNF, indicating the consistency of the TNF-releasing activity of the three different batches.

On a separate group of donor leukocytes, the supernatants remaining after 80% acetonitrile precipitation of the composition were tested. Whereas 0.04 ml of whole batches of the composition (B0222 and B0213) released 175 and 233 pg/ml of TNF, 0.05 ml of the supernatant released 42 and 41 pg/ml about 33% of the activity of the whole composition. Blanks prepared in the same fashion with 80% acetonitrile had no TNF-releasing activity. Thus, TNF-releasing activity in the precipitate was not the result of some residual substances in the buffers used to induce the precipitate.

Whereas in the aforedescribed studies the precipitate and supernatant fractions were tested on leukocytes from different donors, another study was conducted in which the two fractions were tested on leukocytes from the same donors. For the two tested batches, the whole composition at 0.04 ml released between 0 and 41 pg/ml of TNF. In comparison, the precipitate of the same batches of composition at 0.05 ml released 141 and 749 pg/ml of TNF, whereas 0.05 ml of the supernatant fraction released between 6 and 57 pg/ml of TNF. Thus the precipitate contained much of the TNF releasing activity. The supernatant fraction still had some TNF-releasing activity but much less than the precipitate and no more than the whole composition.,

2. Reversed Phase-HPLC Separated Fractions of the Composition of the Invention

As the precipitate of the composition was shown to contain much of the TNF-releasing activity, the profile of the precipitate was examined by C 18 RP-HPLC. FIGS. 18 , 19 and 20 show the RP-HPLC profiles of whole composition, the precipitate and the supernatant, respectively. The precipitate's profile shows principally the early eluting peak of whole composition (FIG. 19 ), whereas the supernatant's profile ( FIG. 20 ) is similar to the whole composition except that the early peak is less intense.

In the next series of experiments, the activity of the precipitate and supernatants were evaluated after their separation by C 18 RP-HPLC. The precipitate and supernatant were collected as 2 pools from RP-HPLC: 2 to 21 min and 21 to 46 min. Earlier studies of fractions of whole composition separated by RP-HPLC indicated that TNF-releasing activity eluted principally in the 2 to 20 min fraction. Table 30 shows the results of testing whole composition and RP-HPLC fractions of the precipitate and supernatant. In this particular experiment, only 10 μl of the whole composition was used and caused no TNF-release. The precipitate pool from 2 to 21 min released TNF, whereas the 21 to 46 min pool did not (Table 30). The supernatant pool from 2 to 21 min also released TNF activity, whereas the 21 to 46 min pool released minimal quantities of TNF (Table 30). Thus for both the precipitate and supernatant, the TNF-releasing activity resides principally in the early eluting fraction from C 18 RP-HPLC.

TABLE 30
TNF Releasing Activity of HPLC Fractions of Precipitate and Supernatants of
Virulizin Prepared in 80% Acetonitrile (2× wash)
	199 Medium	Quantity in	TNF Released (pg/ml)	Osmolarity
Sample Tested	Used	Well	Total	−LPS	(mOsm)
LPS	1X	50	ng	99 ± 43	0	307
Virulizin: B0222	80% 1X	10	μl	83 ± 33	0	289
Whole
HPLC Precipitate:
2:00-21:36	70% 1X	100	μl	153 ± 43	54	306
	70% 1X	50	μl	50 ± 14	0	289
21:36-46:28	1X	100	μl	97 ± 29	0	306
	1X	50	μl	103 ± 43	3	306
HPLC Supernatant:
2:00-21:35	1X	100	μl	192 ± 55	92	330
	1X	50	μl	183 ± 72	83	320
21:35-46:30	1X	100	μl	65 ± 21	0	311
	1X	50	μl	142 ± 40	43	309
Virulizin: B0213	90% 1X	10	μl	93 ± 36	0	296
Whole
HPLC Precipitate:
2:00-21:12	70% 1X	100	μl	165 ± 52	66	310
	70% 1X	50	μl	48 ± 15	0	287
21:12-46:12	1X	100	μl	88 ± 31	0	307
	1X	50	μl	101 ± 38	1	307
HPLC Supernatant:
2:00-21:15	1X	100	μl	193 ± 76	93	317
	1X	50	μl	126 ± 43	26	306
21:15-46:20	1X	100	μl	59 ± 24	0	309
	1X	50	μl	126 ± 32	26	312
Note:
1. Number of patients tested: 5.
2. Total TNF released is corrected for TNF release by 1X 199 Medium.
3. Osmolarities not corrected for 1X 199 Medium (309 mOsm); standard errors are very low and, therefore, not reported.
4. Reconstitution: precipates in Type 1 water, supernatants in PBS buffer.
5. LPS volume added to wells was 10 μl.
6. Wells contained a total volume of 1000 μl.

The results suggest that the TNF-releasing substance in the precipitate and supernatant are likely closely related molecules, if not identical, with the only difference, if any, perhaps being the degree of solubility in 80% acetonitrile.

In another experiment the TNF-releasing activity of the two RP-HPLC pools of the precipitate were examined for three batches of the composition. Table 31 shows again that for three different batches (B0222, B0221, and B0213), the TNF-releasing activity is principally in the pool 2 to 21 min. Thus different batches are consistent.

TABLE 31
TNF Releasing Activity of HPLC Precipitate Fractions of Virulizin Prepared in 80%
Acetonitrile
	199 Medium	Quantity in	TNF Released (pg/ml)	Osmolarity
Sample Tested	Used	Well	Total	−LPS	(mOsm)
LPS (1)	1X	50	ng	136 ± 62	0	299
LPS (1)	1X	50	ng	81 ± 25	0	306
Virulizin: B0222	1X	40	μl	734 ± 276	659	329
Whole		10	μl	119 ± 11	38	303
HPLC Precipitate:
2:00-21:25 min	70% 1X	200	μl	20 ± 4	0	354
		100	μl	155 ± 32	74	307
21:25-46:20	1X	200	μl	67 ± 39	0	313
		100	μl	78 ± 10	0	310
Virulizin: B0221	90% 1X	40	μl	626 ± 90	490	315
Whole		10	μl	62 ± 31	0	290
HPLC Precipitate:
2:00-21.45 min	70% 1X	200	μl	20 ± 7	0	365
		100	μl	170 ± 6	90	329
21:45-46:50 min.	1X	200	μl	45 ± 5	0	336
		100	μl	73 ± 15	0	311
Virulizin: B0213	90% 1X	40	μl	620 ± 123	484	315
Whole		10	μl	80 ± 17	0	290
HPLC Precipitate:
2:00-21:30 min	70% 1X	200	μl	6 ± 4	0	393
		100	[2l	153 ± 45	72	314
21:30-36:40 min	1X	200	μl	182 ± 96	47	312
		100	μl	78 ± 20	0	310
Note:
1. For 100 & 10 μl samples; 3 patients tested for these samples.
2. For 200 & 40 μl samples; 4 patients tested for these samples.
3. Total TNF released is corrected for TNF release by 199 Medium 1X.
4. 199 Medium 1X @ 100 & 10 μl - 306 mOsm, @ 200 & 40 μl - 305 mOsm.
5. Osmolarities are averages and not corrected for 199 Medium 1X; standard errors are not reported as values are very low.
6. LPS volume added to wells was 10 μl.
7. Wells contained a total volume of 1000 μl.
8. Sample volumes are equivalent.
9. Reconstitution: first fractions is double distilled water, second in PBS buffer.

In a further experiment, the effect of washing the precipitate with 80% acetonitrile was examined. The point of the experiment was to prove that the TNF releasing activity was not being simply trapped. Whole composition, 0.04 ml, released 325 pg/ml of TNF. The precipitate pool from 2 to 24 min at 0.1 ml released 324 pg/ml of TNF and at 0.05 ml, 3 pg/ml. The pool from 24 min to 46 min released no TNF. Likewise the supernatant pool from 2 to 24 min released TNF at 0.1 and 0.05 ml, whereas the pool from 24 to 46 min had some, but considerably less, TNF-releasing activity on RP-HPLC.

To be certain that the handling of RP-HPLC isolates of the composition were not responsible for the presence of TNF-releasing activity, samples were prepared in the same fashion but without the composition. RP-HPLC profiles of PBS and H 2 O blanks, their precipitates and supernatants were essentially free of any peaks.

Using mononuclear cells from identical donors, the samples were tested on leukocytes from the same donors. Whole composition and the precipitate eluting from 2 to 24 min released TNF, whereas PBS, H 2 O or their precipitates separated on RP-HPLC had no TNF releasing activity in the pool from 2 to 23 min. Thus, the precipitate eluting from 2 to 24 min causes specific TNF-release.

The precipitate pool of the composition eluting from 24 to 46 min released no TNF. The controls of water and PBS showed release of 114 and 40 pg/ml, respectively. Thus there was no specific release of TNF from the precipitate pool 24 to 46 min.

The supernatant fraction pool 2 to 24 min and 24 to 46 min released 82 and 68 pg/ml of TNF, respectively. The water and PBS blank pools from RP-HPLC released some TNF activity. The water pool 2 to 25 min and 25 to 46 min released 149 and 216 pg/ml of TNF respectively, and the PBS pools released 0 and 126 pg/ml respectively.

Thus, both the precipitated and supernatant fraction had TNF-releasing activity. RP-HPLC separation of the TNF-releasing activity showed that both eluted early from RP-HPLC, suggesting that the active components are physically very similar if not identical.

Table 32 provides a summary result of further testing, for 80% acetonitrile precipitates and supernatants after RP-HPLC separation, minus the activity in similarly prepared blank samples.

TABLE 32
Releasing activity of Virulizin less release by reconstitution solutions.
	199 Media	Quantity	Actual Released (pg/ml)	Osmolarity
Sample Tested	Used	in Well	TNF-a	GM-CSF	IL-1β	(mOsm)
Virulizin: B0222	80% 1X	40 μl	178	136	142	310
Whole
HPLC Precipitate (min):
2:20-24:10	70% 1X	100 μl	75	13	18	314
24:10-46:20	70% 1X	100 μl	0	16	0	292
HPLC Supernatant (min):
2:00-23:55	1X	100 μl	82	86	0	336
23:55-46:20	1X	100 μl	0	45	29	316
Note:
1. Data derived from Virulizin Sumtable Table 24.4.
2. Virulizin reconstitutition: 1st precipitate fraction in Type 1 water, all other fractions in PBS.
3. Whole amounts are unreconstituted and therefore not corrected.
4. Sample volumes are equivalent.

Whole composition releases TNF, GM-CSF and IL-1β in vitro from mononuclear cells at 24 hrs. The 80% acetonitrile precipitate contains the same releasing activity and most elutes in the early fraction from RP-HPLC. The supernatant fraction retains releasing activity and most elutes in the early fraction from RP-HPLC. The supernatant fraction retains releasing activity, but it also elutes in the early eluting RP-HPLC fraction for TNF and GM-CSF. The results suggest that likely the same component releases the three cytokines. For GM-CSF and IL-1β, the fact that some releasing activity elutes in the late fraction from RP-HPLC suggest that there may be another substance in the composition that can act on monocytes to release GM-CSF and IL-1β.

Physicochemical Analysis SDS Gel Electrophoresis

Having identified that TNF, IL-1β and GM-CSF releasing activity can be precipitated, in part, by 80% acetonitrile and that much of the releasing activity elutes early from C 18 RP-HPLC, the physicochemical properties of the precipitate fraction have been studied and compared to the whole composition and supernatant fraction of the composition.

FIG. 21 shows an SDS gel electrophoresis of whole composition and precipitates and supernatants of the composition. In all three instances, the composition runs near the SDS front, indicating a low molecular weight. The smallest standard used was 14,400 daltons.

Molecular Sieve HPLC

The molecular size of the composition was also examined by determining its time of elution from a molecular sieve HPLC column. The elution times of whole composition, precipitate and supernatant compared to standards. All three eluted later than insulin, which eluted at 24.5 min. Once again, physicochemical analysis indicates a mol. wt. less than 2,400 daltons.

Hydrophilic (Polyhydroxyethyl) HPLC

The TNF-releasing component elutes early. Thus a column with the opposite effect was chosen, a hydrophilic column in the presence of organic solvents. The ideal eluting conditions for the polyhydroxyethyl column is 80% acetonitrile. However, as indicated in the prior Example, some of the substances in the preparation precipitated at this concentration. Consequently, the composition was analyzed at a low concentration of acetonitrile where the column functions mostly as a molecular sieve column. FIGS. 22 and 23 show the profile of whole supernatant and precipitate. The front sheet summarizes the elution time for the different peaks. The elution times indicate the active componet of the composition has a low molecular weight.

Amino Acid Analysis and Sequencing of the Precipitated Component

Two samples were submitted for protein analysis by amino acid compositional analysis before and after acid hydrolysis: the acetonitrile precipitate and 2-20 min RP-HPLC eluted pool of the precipitate. The two samples were very similar by comparison of amino acid content before and post acid hydrolysis. There are, however, significant differences between the amino acid composition (post acid hydrolysis), and the free amino acid content which suggests that peptide bonds were hydrolyzed. The composition of the samples were peculiar in that they were very rich in glycine plus glutamate/glutamine.

From the foregoing, it may speculated that at least one of the active components is proteineous. Analysis reveals potentially significant quantities of unidentified ninhydrin positive (most likely amino acid compounds, but other compounds may yield a response) components that appear to be stable to acid hydrolysis. The principal amino acids per 1000 residues in the sample are Asx (asparagine) 143, Thr (Threonine) 31, Glx (glutamate) 381, Gly (glycine) 187, and Ala (alanine) 170.

A comparison was made between free and released—(acid hydrolysis) amino acids. This shows the following amino acids per 1000 residues: Asx (asparagine) 51, Threonine 8.6, Serine 18, Glx (glutamine) 375, Pro (proline) 17, Glycine-(Glyj 429 and Alanine 86.

The ratio of Free/1000 is as follows:

Released/1000
Asx 4.09
Thr 2.03
Ser 1.65
Glx 1.18
Gly 0.37
Ala 1.69

There are 5 unidentified (ninhydrin positive) components. Highly speculative assignments are cysteic acid, glucosaminic acid and sarcosine. Also there may be methionine sulfoxide and methionine sulfone. The major unidentified ninhydrin component appears to be free components in the sample.

EXAMPLE 12

Release of IL-1β and IL-8

Table 33 shows that the composition of the invention stimulates human mononuclear cells in culture to release IL-1β and the 80% acetonitrile precipitate of the composition releases more than the remaining supernatant. Whereas the whole composition does not stimulate IL-8 release, fractionated composition seems to release some IL-8. The results shown in Table 33 are minus the release of IL-1β and IL-8 with mock samples.

TABLE 33
Releasing activity of Virulizin less release by reconstitution solutions.
	199 Media	Quantity	Actual Released	Osmolarity
Sample Tested	Used	in Well	IL-1β (pg/ml)	IL-8 (ng/ml)	(mOsm)
Virulizin: B0222	80% 1X	40 μl	171	0	304
Whole
Precipitate	70% 1X	100 μl	160	71	292
Supernatant	90% 1X	100 μl	17	87	335
Note:
1. Number of patients tested: 5.
2. Reconstitution: precipitate in Type 1 water, supernatant in PBS buffer.
3. Samples are corrected for release by 1X 199 Medium and LPS: 154, 1065 pg/ml, and 68, 294 ng/ml, respectively, for IL-16 and IL-8.
4. Osmolarities for media and LPS are: 311 and 309, respectively.
5. LPS volume ADDED TO wells: 10 μl.
6. Total volume of wells: 1000 μl.
7. Sample volumes are equivalent.

Physicochemical Characteristics

The composition and its precipitate and supernatant were separated by ion-exchange HPLC. Both by AX300 (anion exchange) chromatography and by CMX 300 (cation exchange) chromatography, there was no significant separation of components. Hydrophobic reverse phase chromatography did not separate the peaks.

Capillary Electrophoresis

The precipitate was analyzed by capillary electrophoresis. At high pH, a W absorbing peak was observed at 190 nm but completely disappeared at 200 nm. There were no significant peaks at 214 nm Uv absorption.

Free amino acid are not visualized unless they are derivatized. It is thought that the W peak at 190 nm is likely a salt.

EXAMPLE 13

The composition was evaluated for stimulatory activity in the following 3 indicator systems: 1) Stimulation of lymphocyte DNA synthesis; 2) Induction of lymphocyte-mediated cytotoxic function; and 3) Induction of monocyte/macrophage-mediated cytotoxic function. These tests were chosen for the screen because they measure immunological functions which have been shown to be associated with different clinical parameters in patients with malignant disease. These indicators of immune function also can be modulated in cancer patients who are treated with different biological response modifying agents such as interferon or interleukin-2. The results of the initial screening procedure are presented below.

1) Stimulation of Lymphocyte DNA synthesis: comparison with an optimal stimulating concentration of phytohemaaalutinin (PHA)

Stimulant          Counts per Minute
Medium             374
PHA                125,817
Composition (#222) 1,116
Composition (1:10) 1,021
Composition (1:50) 649

Results: Unlike the prototypic mitogen, PHA, the Composition does not stimulate lymphocytes to undergo blastogenesis and cell division.

2) Stimulation of Lymphocyte-mediated Cytotoxic Function: comparison with an optimal stimulating concentration of Interleukin-2 (IL-2)

Stimulant          Lytic Units
Medium             30.8
IL-2               472.5
Composition (neat) 48.1
Composition (1:10) 33.3
Composition (1:50) 44.8

Results: Unlike the prototypic stimulator of lymphocyte cytotoxic function, Interleukin-2, the composition does not elicit lymphocyte cytotoxicity.

3) Stimulation of Monocyte-Mediated Cytotoxic Function by the Composition: Comparison with Gamma Interferon & Endotoxin (γ-IFN+LPS)

Stimulant
(E/T = 20/1)       % Cytotoxicity
Medium             4.3
IFN + LPS          24.4
Composition (neat) 19.7
Composition (1:10) 20.0
Composition (1:50) 11.5

Results: The composition is capable of stimulating peripheral blood monocytes to express tumoricidal function in a dose dependent manner. The magnitude of stimulation is comparable to that elicited by the prototypic macrophage activator combination of γIFN+LPS. It is important to recognize that the action of the composition in these in vitro assays did not require the addition of endotoxin as in the case with any other macrophage activators. If the composition is free of endotoxin contamination, its biological activity in this assay of macrophage activation would be considered biologically significant.

Monocyte/Macrophage Studies With the Composition

Because the screening procedures demonstrated that the composition does not stimulate lymphocyte functions but can stimulate monocyte functions, subsequent studies were aimed at further characterization of the monocyte/macrophage stimulatory activities of this compound. A number of comparative studies aimed at determining the dose response characteristics of the composition in stimulating monocyte/macrophage tumoricidal function, were performed as well as testing different batches of the compound. The main emphasis of the studies was to test the capacity of the composition to simulate tumoricidal function in monocytes and macrophages from different anatomical sites of cancer patients. The central hypothesis guiding these studies is that the therapeutic efficacy of any biological stimulator will depend, in large part, on its ability to elicit tumoricidal function in environments which contain malignant disease. That could come about by direct stimulation of resident immune cells in tumor microenvironments. Alternatively, this could come about by stimulation of circulating immune cells if those cells were then able to home on sites of malignant disease and to function in that environment. For these investigations, the following were relied upon: 1) peripheral blood monocytes from cancer patients and control subjects; 2) alveolar macrophages from lung cancer patients and control patients with non-malignant lung diseases; and 3) Peritoneal macrophages from patients with gynecological malignancies.

1. Dose Response and Different Batch Studies With the Composition

These studies relied on peripheral blood monocytes to test the stimulatory activities of different doses and different batches of the composition. Three batches of the composition were provided for testing. These were designated as batch #s 216, 219 and 222. Each batch of the composition was tested without dilution (neat), a 1:10 dilution and a 1:50 dilution of material. The results are depicted graphically in FIG. 24 .

Results: Batch #222 and 216 stimulate monocyte tumoricidal function, Batch #219 did not. It appeared that #222 was superior to 216 in these preliminary investigations. Batch #222 appears to stimulate equivalent levels of tumoricidal function at the undiluted (neat) and 1:10 dilution concentration with less, but still detectable activity at the 1:50 dilution. Batch #216 gave the greatest stimulation of tumoricidal function at the undiluted (neat) concentration, with less activity at the 1:10 dilution and no detectable activity at the 1:50 dilution. As stated above, Batch #219 did not elicit detectable monocyte tumoricidal function at any concentration tested.

2. Tumoricidal Function in Peripheral Blood Monocytes

Tests have been performed on 4 peripheral blood monocyte samples from control subjects. These tests utilized an optimal stimulating concentration of the composition (1:10 dilution of batch #222) and an optimal stimulating concentration of γ-IFN+LPS. The target cells in these studies were a cultured, NK-insensitive cell line, the Chang Hepatoma.

Stimulant   % Cytotoxicity
(E/T = 20/1)
Medium      5.4 +/− 1
γ-IFN + LPS 18.6 +/− 4
Composition 22.3 +/− 6

A test was also performed on 1 monocyte sample from a patient with cervical cancer. This test was important because the patient's own tumor cells were available to be used as target cells in the assay. As before, this test utilized an optimal stimulating concentration of the composition (1:10 dilution of batch #222) and an optimal stimulating concentration of γ-IFN+LPS. Also, the effector/target cell ratio was reduced to 15/1 to conserve patient tumor cells.

Stimulant   % Cytotoxicity
(E/T = 20/1)
Medium      5.5
γ-IFN + LPS 14.4
Composition 20.9

Results: In the peripheral blood monocytes from control subjects, the composition stimulated monocyte tumoricidal function against the Chang Hepatoma at a level equal to or greater than the level elicited by an optimal stimulating concentration of γIFN+LPS. In the peripheral blood monocytes from a patient with cervical cancer, the composition stimulated tumoricidal function against the patient's own tumor cells at a level which exceeded that elicited by γ-IFN+LPS by >30%.

3. Tumoricidal Function in Peritoneal Macrophages From Patients With Gynecological Malignancies

These tests were performed on peritoneal macrophage samples isolated from lavage fluids of 1 patient with cervical cancer and 1 patient with Ovarian Cancer. These tests were performed with the patient's own tumor cells as target cells in the assay. As before, an optimal stimulating concentration of the composition (1:10 dilution of batch #222) and an optimal stimulating concentration of γ-IFN+LPS were compared. Also, the effector/target cell ratio was reduced to 15/1 to conserve patient tumor cells.

	Stimulant	Cervical Cancer	Ovarian Cancer
	Medium	8.2	0.6
	IFN + LPS	29.8	4.1
	Composition	13.2	8.9
	(1:10)

Results: These test results highlighted the fact that the local tumor environment may be a determinant of the response of immune cells to immunological activators. In this case of cervical cancer, there was no pathological evidence of malignant disease within the peritoneal cavity and the development of tumoricidal function against the autologous tumor was better with γIFN+LPS than the composition. In the patient with ovarian cancer, there was significant tumor in the peritoneal cavity. The response against the patient's own tumor to γIFN+LPS was minimal at best, whereas the response to the composition was greater.

4. Tumoricidal Function in Alveolar Macrophages From Lung Cancer Patients and Control Subjects

These tests were performed on alveolar macrophage samples isolated from broncholveolar lavage fluids of a patient with non-small cell lung cancer and 3 patients with non-malignant diseases of the lung. These tests utilized an optimal stimulating concentration of the composition (1:10 dilution of batch #222) and an optimal stimulating concentration of γ-IFN+LPS. The target cells in these studies were the Chang Hepatoma cells and the effector/target cell ratio was 20/1.

	Stimulant	Cancer Patients	Control
	Medium	2.6 +/− 2	19.5 +/− 4
	γ-IFN + LPS	10.9 +/− 13	1.2 +/− 5
	Composition	5.2 +/− 2	18.6 +/− 8

Results: Alveolar macrophages from lung cancer patients are impaired in their development of tumoricidal function in response to conventional macrophage activators such as γIFN+LPS. These results are consistent with this observation; they show that the tumoricidal function of alveolar macrophages from lung cancer patients is greatly reduced compared to control subjects. They also show that the composition does not activate tumoricidal function in either the alveolar macrophages of lung cancer patients or the alveolar macrophages of control subjects with non-malignant lung diseases.

These preliminary in vitro tests with the composition demonstrate that it is a macrophage activator. The material provided was able to elicit tumoricidal activity in a standard cytotoxicity assay against both an NKinsensitive cell line and against freshly dissociated human tumor cells. The activity elicited was also found to be concentration dependent in these tests. The capacity of the composition to active macrophage tumoricidal function in vitro is comparable to that of the best macrophage activating combination presently available, namely, γIFN+endotoxin. As stated above, the capacity of the composition to elicit this level of tumoricidal function in the absence of endotoxin would be considered important biologically if the material is free of endotoxin contamination.

As has been found for other macrophage activators, the activity of the composition in stimulating macrophage tumoricidal function varies with the source of the macrophages. It appears that the composition is an excellent activator of peripheral blood monocytes being equivalent to γIFN+LPS with normal donors and possibly superior to γIFN+LPS with cancer patient donors. Malignant disease has a significant impact on the development of monocyte tumoricidal function depending on the activator used (Braun et al, 1991). One determinant of the biological activity of different macrophage activators in cancer patients monocytes is the sensitivity of the activator to arachidonic acid metabolism and the secretion, by the cell of prostaglandins. From these initial studies with the composition, it appears that activity elicited with the compound is not sensitive to the inhibitory effects of prostaglandins. If prostaglandin insensitivity can be proven definitively for cancer patient monocytes stimulated with the composition, this would be considered important therapeutically since the effectiveness of many other biological activators is limited by prostaglandins. Preliminary studies with 2 specimens indicate that the composition may have good activity in peritoneal macrophages, particularly when malignant disease is present in the peritoneal cavity.

These preliminary results also illustrate what has been found when comparing the capacity of different activators to stimulate tumoricidal function in peritoneal macrophages of patients with different gynecological malignancies. In those studies, it was found that the presence of malignant disease within the peritoneal cavity influences the responsiveness of the peritoneal macrophages to specific activators. In patients with cervical cancer, malignant disease is not present in the peritoneal cavity in general, and thus, the response of the resident macrophages to γIFN+LPS is normal. When disease is present in the cavity, however, as in the case with ovarian cancer, the response to γIFN+LPS is suppressed. This is related, in part, to changes in the arachidonic acid metabolism of the peritoneal macrophages when malignant disease is present (Braun et al, 1993). The fact that the composition apparently can activate tumoricidal function in peritoneal macrophages from ovarian cancer patients against the patient's own tumor cells may reflect, once again, a mechanism for activation which is independent of the arachidonic acid metabolic pathway.

On the other hand, the composition clearly does not activate alveolar macrophages to become tumoricidal whether malignant disease is present in the lung or not. Alveolar macrophages from lung cancer patients have been found to be inhibited significantly in their development of tumoricidal function when compared to either peripheral blood monocytes from the same patients or to control alveolar macrophages from patients with non-malignant lung diseases (Siziopikou et al., 1991). Thus, the lack of activity of the composition in this case is not surprising.

EXAMPLE 14

The development of tumoricidal function in response to the composition of the invention and other macrophage activators was investigated in peripheral blood monocytes and peritoneal macrophages from patients with gynecological diseases. More particularly, the patient population consisted of 7 patients, 3 with benign disease and 4 with malignant disease (2 ovarian cancers, 1 endometrial cancer, and 1 cervical cancer). Samples were removed from patients at the time of surgical procedure. Preparations containing peripheral blood monocytes were isolated from blood samples using the procedure set out in Braun et al. Cancer Immunol. Immunother 32:55-61, 1990 and preparations containing peritoneal macrophages were isolated as set out in Braun et al., Cancer Research 53:3362, 1993. Tumor cell cytotoxicity in response to the composition of the invention (1:10 dilution of stock batch 222) and other activators namely gamma interferon (100 U/ml), interleukin-12 (500 U/ml), and monocyte-CSF (500 U/ml) was assessed using the monocyte cytotoxicity assay described in Braun et al., Cancer Immunol. Immunother 32:55-61, 1990.

The results as shown in Table 34, demonstrate that the composition of the invention stimulates tumoricidal function in both the peripheral blood monocytes and the peritoneal macrophages from patients with malignant and non-malignant gynecological diseases. The tumor cytotoxicity elicited by the composition of the invention is equal to or greater than that elicited by the other biological stimulators which were tested.

TABLE 34
The Development of Tumoricidal Function in Response to the
Composition of the invention and other Macrophage Activators
in Peripheral Blood Monocytes and Peritoneal Macrophages from
Patients with Gynecological Diseases
(3 benign disease, 4 malignant disease)
% Tumor Cytotoxicity (+/− S.E.)
at Monocyte/Tumor Cell ratio - 15/1
		Peripheral	Peritoneal
	Activator	Blood	Macrophage
	Medium	8.6 ± 3	3.1 ± 1
	Gamma Interferon	18.3 ± 2	9.5 ± 1
	Interleukin-12	26.0 ± 4	8.5 ± 2
	Monocyte-CSF	16.0 ± 2	7.0 ± 2
	Composition of the	23.0 ± 6	12.5 ± 2
	Invention
	(Virulizin)

EXAMPLE 15

The effect of indomethacin, a prostaglandin synthesis inhibitor, on the development of tumoricidal function in response to the composition of the invention and other macrophage activators in peripheral blood monocytes from cancer patients was also investigated. Samples from the Patients with malignant disease in Example 14 were tested using the assay system as described in Example 14 with the exception that indomethacin (up to 5 ng/ml) was simultaneously added with the composition of the invention, interleukin-12 (500 U/ml), and monocyte-CSF (500 U/ml).

The results as shown in Table 35 indicate that indomethacin augments cytotoxicity in response to IFNa, GM-CSF and N-CSF. Thus, the development of tumoricidal function in response to IFN-y, GM-CSF, and M-CSF was regulated by an indomethacin-sensitive function. In contrast, the development of tumoridical function in response to Phorbol Ester (PNA), IL-12 and the composition of the invention was not regulated by an indomethacinsensitive function i.e. indomethacin did not augment cytotoxicity in response to the composition of the invention, IL-12 and PNA.

TABLE 35
The effect of Indomethacin, a Prostaglandin Synthesis Inhibitor,
on the Development of Tumoricidal Function in Response to the
Composition of the Invention and other Macrophage Activators
in Peripheral Blood Monocytes from Cancer Patients.
	Activation Conditions	# donors	% cytotoxicity
	*IFN-γ	23	11.9 ± 9
	*IFN-γ + Indomethacin		25.2 ± 17
	*GM-CSF	10	7.8 ± 6
	*GM-CSF + Indomethacin		17.8 ± 8
	*PMA	6	27.3 ± 14
	*PMA + Indomethacin		22.0 ± 17
	IL-12	3	24.7 ± 5
	IL-12 + Indomethacin		25.6 ± 6
	M-CSF	3	14.1 ± 3
	M-CSF + Indomethacin		19.0 ± 3
	Composition	4	18.7 ± 6
	(Virulizin)
	Composition		16.4 ± 6
	(Virulizin) +
	Indomethacin

EXAMPLE 16

The effect of prostaglandin E 2 on the development of tumoricidal function in response to the composition of the invention in the presence of indomethacin was investigated. The subject population consisted of one normal and eight patients (one patient with a pancreatic tumor, two patients with head and neck tumors, one with endometriosis, and four with HIV). Preparations containing peripheral blood monocytes were isolated from blood samples from the patients using the procedure set out in Braun et al. Cancer Immunol. Immunother 32:55-61, 1990. Tumor cell cytotoxicity in. response to the composition of the invention (1:10 dilution of stock batch 222) and indomethacin (up to 5˜g/ml), with or without PGE 2 (10 8 M), was assessed using the monocyte cytotoxicity assay described in Braun et al., Cancer Immunol. Immunother 32:55-61, 1990.

The results in Table 36 show that 36 pathophysiological levels of PGE 2 (10 8 M) failed to suppress the level of tumoricidal function which developed in response to the composition of the invention. This is in contrast to the capacity of PGE 2 to suppress tumoricidal function in monocytes stimulated with γ-interferon (Braun et al, Cancer Research 53:3362, 1993).

TABLE 36
The Effect of Prostaglandin E 2 on the Development of
Tumoricidal Function in Response to the Composition of the
Invention in the Presence of Indomethacin.
% Tumor Cytotoxicity at Monocyte/Tumor Cell ration = 15/1
			Composition
		Composition	(Virulizin) +
	Composition	(Virulizin) +	Indomethacin +
Diagnosis	(Virulizin)	Indomethacin	PGE 2
Normal	19	20	27
Pancreatic	15	14	22
HNSCC	9	8	12
HNSCC	11	3	12
Endometriosis	37	37	n.d.
HIV	6	7	8
HIV	15	12	19
HIV	21	16	20
HIV	23	22	n.d.

EXAMPLE 17

The development of tumoridical function against autologous tumor cells in monocytes stimulated with the composition of the invention was investigated. Preparations containing peripheral-blood monocytes were isolated from blood samples from 6 patients (three ovarian cancers, one endometrial cancer, one cervical cancer and one ENT cancer) using the procedure set out in Braun et al., 1990. Tumor cell cytotoxicity in response to the composition of the invention (1:10 dilution of stock batch 222) and indomethacin (up to 5˜g/ml), with or without PGE(10 8 N) was assessed using the monocyte cytotoxicity assay described in Braun et al., 1990, with the exception that the patient's tumor cells were used in place of the Chang hepatoma cells. The patient's tumor cells were treated with collagenase and DNase, single cell preparations were prepared, and the cells were labelled as described in Braun et al. 1990.

The results shown in Table 37 demonstrate that the composition of the invention is capable of activating the patient's own monocytes to kill the patient's tumor. The composition of the inventon is at least as effective as the standard biological activators which are currently being used.

TABLE 37
The Development of Tumoricidal Function against Autologous
Tumor Cells in Monocytes Stimulated with the Composition
of the Invention (Virulizin)
			% Tumor
		Culture	Cytotoxicity
	Diagnosis	Conditions	(E/T - 15/1)
	Ovarian Cancer	Medium	2
		Composition (Virulizin)	11
	Ovarian Cancer	Medium	1
		γ-Interferon + LPS	4
		Composition (Virulizin)	9
	Ovarian Cancer	Medium	0
		γ-Interferon + LPS	14
		Composition (Virulizin)	11
	Endometrial Cancer	Medium	6
		γ-Interferon + LPS	14
		Composition (Virulizin)	21
	Cervical Cancer	Medium	8
		γ-Interferon + LPS	30
		Composition (Virulizin)	13
	ENT Cancer	Medium	11
		γ-Interferon + LPS	12
		Composition (Virulizin)	25

The experimental results in Examples 14 to 17 indicate that the composition of the invention is capable of activating monocytes to express tumoricidal function, and it is at least as effective as other activators currently being used in the clinical setting; it works in the blood with peritoneal macrophages; and, it appears to not be subject to the inhibitory effects of prostaglandins, which is one of the principle forms of immunosuppression in patients. The experimental data also supports the utility of the composition in the treatment of peritoneal and gynecological malignancies.

EXAMPLE 18

Early Toxicity Studies

Toxicity studies were conducted on a variety of animal species. The studies are summarized in Table 38. All animals were assessed on the basis of daily clinical observation while receiving the injections on days 14, 21 and 30 thereafter. No adverse effects were noted throughout the period that injections were administered or during the follow-up period (one month for all species except the dogs which were followed for 4 months).

TABLE 38
Summary of Early Toxicity Studies
Animal	Quantity	Dose
White Mice	100	0.2 ml i.m. at three day intervals four times
Male Wistar	100	2.0 ml i.m. at three day intervals four times
Rats
Golden	60	1.5 ml i.m. at four day intervals four times
Hamsters
Guinea Pigs	60	3.0 ml at three day intervals four times
Rabbits	15	5.0 ml i.m. at three day intervals four times
Cats	10	3.0 ml i.m. at three day intervals six times
Dogs	12	2 ml/kg i.m. given once - observed for
		four months**
Hematologic data collected every third day for the first 30 days and once monthly thereafter.

A toxicity study was conducted to determine the effect of a single large intramuscular dose of the composition. Thirteen rats received a single intramuscular dose of 5 ml/kg of the composition. Three rats were observed for 7 days. Ten rats were observed for 14 days followed by euthanasia and necropsy. No symptoms of toxicity were observed in either group and no gross pathologic findings were observed in the animals that were necropsied. Based on these observations the LD 50 for intramuscular administration of the composition in rats was determined to be greater than 5 ml/kg. Table 39 summarizes these results.

TABLE 39
Estimation of LD 50 in Sprague-Dawley Rats
				Route of
Animal	Quantity	Admin.	Dose	Units/kg*	LD50
Sprague	3 male	i.m.	5 ml/kg	52.5	>5
Dawley Rats
	10	i.m.	5 ml/kg	52.5	>5 ml/kg
	(5 male/
	5 female)
*Units calculated on the expected range of bioactivity of Lot #B0201 measured at 10.5 units/ml.

Toxicity Trial In Dogs

In a study conducted by the Ontario Veterinary College, the composition was administered to two mixed breed dogs. The protocol is summarized in Table 40. In each case one dose was given in the right leg and the second dose 7 days later was given in the left rear leg. Both dogs were observed for 14 days after the first injection. Appetite, activity, temperature, pulse rate, respiratory rate were monitored twice daily throughout the study. Routine urinalyses, hematology and serum chemistry profiles were performed, pretreatment and 24 hours, 72 hours, 7 days and 14 days after the first injection. Neither animal showed signs of pain associated with either injection. There was no evidence of anaphylaxis associated with the second injection. No abnormalities or changes in physical or laboratory parameters were observed that could be attributed to the drug. The drug appeared to be well tolerated by healthy dogs.

TABLE 40
Toxicity Study in Dogs
	Age and	Dose 1 Units	Dose 2 Units	Dose
Animal	Weight	Calculated*	Calculated*	Interval
Male	Adult	5.5 ml i.m.	0.6 ml i.m.	7 days
Mixed Breed	5 kg	28.6-50.6 Units	3.1-5.5 Units
Female	6 months	12.5 ml i.m.	1.3 ml i.m.	7 days
Mixed Breed	13 kg	65.0-115.0 Units	6.8-12.0 Units

Treatment of Animals With Malignant Neoplasmas

The composition of the invention was used clinically in a veterinary hospital for the treatment of various malignant tumors in companion animals. Eleven cats and ten dogs with advanced neoplastic disease that was not responding to conventional therapy were treated with the composition given intramuscularly in weekly doses. Table 41 summarizes the individual clinical cases in this study. The number of injections ranged from 2 to 69, with volumes up to 7.5 ml given into a single intramuscular site. Protocols of weekly injections allowed for examinations and careful monitoring of the individual cases with diagnostic tests determined individually for each case. The clinician noted that there was no local irritation nor severe allergic reactions, including anaphylaxis. The clinician and the owners of the animals did not observe any systemic adverse reactions. The investigators noted some clinical improvements consisting of minor reductions, improved appetite and activity levels, significant weight gain in a few animals and a decrease in pain and/or discomfort.

TABLE 41
Table 5-5 Summary of Results from Treatment of Animals with Malignant Neoplasms
No.	Name-Age	Species-Sex	Diagnosis	From-To	Injections	Surgeries	Results
01	Bandit-13	Canine-M/n	Orinasal	01.31.87-05.19.87	16	3	Minor Partial Response
			Fibrosarcoma				Progressive Disease
							Euthanasia
02	Bob-5	Feline-M/n	Focal Osseous	04.02.87-08.10.87	18	11	FIrst recurrence 16
			Metaplasia with				months all complete
			Osteosarcomatous				response
			development,
			Spindle Cell
			Sarcoma, Feline	11.08.88-02.21.90	69		Response with
			Fibrosarcoma,				subsequent treatment
			Squamous,				evolving to Progressive
			Recurrent Spindle Cell				Disease (tumor became
			Sarcoma - invasive				Rapidly invasive)
			(necropsy diagnosis)				Euthanasia
03	J. D.-7	Canine-M/n	Oral Amelanotic	03.02.02.03.87	26	3	Complete Response.
			Melanoma,	08.24.87	20		Currently Asymptomatic
			Benign Papilloma	12.14.87-04.05.88	14		(4 months)
			Recurrent Round	04.04.89-08.01.89	17
			Cell Sarcoma	11.09.89-11.30.89	15
				241290	4
				25029189	94
04	Mimi-7	Canine-F/s	Invasive	02.27.87-08.10.87	22	5	Stable (No Change),
			Fibrosarcoma recurrent				Limb amputation.
							No recurrence
05	Goliath-17	Feline-M/n	Malignant Melanoma	04.02.87-09.14.87	22	3	Initial Major Partial
			Fibrosarcoma				Response. Subsequent
				11.30.87-07.25.88	31		Minor Partial Response.
							Progressive Disease
							Euthanasia.
06	Diablo-15	Feline-M/n	Malignant	05.28.89-06.29.87	5	1	Initial Minor Response
			Squamous Cell				Then Progressive Disease
			Carcinoma				Euthanasia
07	Oliver-10	Canine-M/n	Malignant Round	02.24.89-05.30.89	12	1	Complete Response.
			Cell Sarcoma				Asymptomatic 1 year.
08	Karu-7	Canine-M/n	Mucinous Intestinal	06.01.87-09.14.87	14	1	Minor Partial Response.
			Carcinoma-Metastatic				Then Progressive Disease.
							Euthanasia.
09	Puppy-12	Feline-M/n	Ceruminous	07.22.88-10.04.88	10	1	Minor Partial Response.
			Gland Adenocarcinoma				Then Progressive Disease.
							Died 10.10.87
10	Grandpa-16	Feline-M/n	Anaplastic	01.17.89-03.20.89	9	2	Minor Partial Response.
			Neoplasma High Grade				Then Progressive Disease.
			Malignancy				Euthanasia 03.26.87
11	Sam-7	Feline-F/s	Mediastinal	11.02.87-12.21.87	7	0	Minor Partial Response.
			Lymphoma				Then Progressive Disease.
							Died 12.21.87
12	Pete-3	Feline-M/n	Acute Feline	04.13.87-05.11.87	5	0	Transient Minor Partial
			Leukemia				Response. Then
							Progressive Disease. Died
							05.05.87
13	Midnight-8	Feline-M/n	Feline Leukemia	11.24.87-01.01.88	2	0	Progressive Disease
							Euthanasia 01.07.88
14	Stormy-10	Canine-M/n	Amelanotic	03.02.87-07.04.87			Complete Response,
			Melanoma				Recurrence
							After Nine Months,
							Progressive Disease
15	Penny-10	Canine-F	Malignant	03.02.87-07.08.87			Partial Response(?)
			Melanoma				Progressive Disease
16	Muky-5	Feline-F	Anaplastic	02.09.87-06.08.87	6	3	Complete Response,
			Carcinoma				Recurrence After Three
							Months Stable
17	George-10	Feline-M	Malignant	03.02.87-08.04.87	15	1	Minor Partial Response,
			Melanoma				No Change After 26
							Months Stable
18	Simon-13	Canine-M	Benign	04.02.87-08.31.87	10	1	Minor Partial Response
			Prostatic
			Hyperplasia
19	Tequila-14	Canine-F/s	Malignant	11.24.8-08.09.90	35	1	Minor Partial Response
			Intestinal				Euthanasia 08.09.90.
			Adenocarcinoma
20	Sheba-12	Canine-F/s	Invasive	12.06.89-06.28.90	25	1	Initial Minor Partial
			Osteosarcoma				Response. Then
			Skull				Progressive Disease
							Euthanasia
21	Mesha-14	Feline-F/s	Osteosarcoma	02.07.89-05.06.89	13	2	Limb Amputation.
							Complete response. No
							recurrence or metastases 1 yr.

The clinical results are summarized as follows: Six animals (3/10 canines and 3/11 felines) experienced complete response. One animal (1/11 felines) had initial major partial response. Eleven animals (5/10 canines and 6/11 feline) experienced minor partial response. One animal (1/10 canines) remained stable and one animal (1/11 felines) did not respond. Table 42 provides definitions of each treatment. The clinical experience in animals suggested a potential role for the composition in the treatment of malignant neoplasms.

TABLE 42
Definition of Treatment Responses
Response            Definition
Complete Response   Disappearance of all clinical evidence of active
                    tumors. The patient must be free of all known
                    disease as determined by two observations not less
                    than four weeks apart.
Partial Response
Major               Where there is a greater than 50% reduction in the
                    sum of the product of the perpendicular dimension
                    of all measurable tumor with no new lesions
                    appearing elsewhere.
Minor               Where there is a 25-50% shrinkage in the sum
                    of the products of the perpendicular diameters of all
                    measurable tumors; or subjective responses such as
                    improvement in performance status, appetite and
                    feeling of well being; or tumor necrosis or lysis
                    as seen on ultrasound, x-rays, or changes in
                    consistency and character of the tumors suggesting
                    a decrease in adhesions and an increase in tumor
                    mobility.
Stable Disease      Less than 25% increase or decrease in the size of
                    one or more measurable lesions without tumoral
                    lysis, or appearance of new lesions.
Progressive Disease Increase of 25% in the size of one or more
                    measurable lesions without tumoral lysis, or
                    appearance of new lesions.

EXAMPLE 19

Preliminary Clinical Trials

Patients with untreatable tumors were treated with 0.11 ml per kilogram of the composition of the invention as prepared in accordance with the methods set out in Example 1. The composition was given intramuscularly every three to five days. Of the 58 patients treated there was absolutely no significant toxicity. In the 37 evaluable patients three patients had minor responses i.e. tumor shrinkage of between 25 and 50 percent, and five patients had stable disease of at least eight weeks in duration. The most interesting results were in the pancreatic patients who seemed to have the most encouraging results. Of the seven patients one patient had the disease stabilization for a full 11 months. And a second patient with extremely advanced disease had disease stabilization for four months. It was based on these results that carcinoma of the pancreas was selected for a basic study of the composition of the invention. The objective of the study was to determine the safety and efficacy of the composition in this group of patients.

The treatment consisted of the composition of the invention 0.11 ml per kilogram with a minimum dose of 7.5 ml given as a single deep intramuscular injection to the gluteus maximus. Patients received the treatment three times weekly during the first week and then twice a week until disease progression. In all, 22 patients were enrolled in this study and all were evaluable for toxicity. Only 17 patients were evaluable for efficacy. With a total of 570 injections there was no toxicity of any kind reported in either local or systemic. There were also no objective responses. Six patients however had stable disease for three months or longer but the rest of the patients progressed within the first three months. The median survival was eight months from the time of diagnosis. There was a median survival of four months from the first injection. There were three patients who had stable disease for longer than six months. One patient, a 75 year old man relapsed with liver metastasis 18 months after a liberal procedure. He remained absolutely stable on the composition for eight months before progressing. A 71 year old woman with unresectable disease remained stable for at least ten months and continued to work full time. A 64 year old woman who relapsed regionally four months after the procedure was stable for at least eight months.

A fourth patient who had inoperable carcinoma of the pancreas and could not be enrolled in the study because tissue could not be obtained for a pathologic diagnosis, was stable for almost a year although his tumor progressed despite higher doses of the composition. In summary, the composition has no site of toxic activity against pancreatic cancer at this dosage schedule. There was a suggestion of temporary anti-proliferative activity in a minority of cases with this disease using the composition.

EXAMPLE 20

Pancreatic Cancer Clinical Studies

A Phase II trial with the composition of the invention was begun for patients with measurable, biopsy-proven pancreatic cancer. The composition was administered as a 7.5 ml (0.11 ml/kg) intramuscular injection 3 times weekly for 1 week then twice weekly until disease progression. Details of the study are set out below.

Method

Treatment consisted of the composition prepared as in Example 1, 0.11 ml/kg (minimum dose 7.5 ml) administered with a single deep intramuscular injection to the gluteus maximus, alternating buttocks with each dose. Patients received 3 injections during the first week followed by twice-weekly injections until tumor progression.

Response was defined using standard criteria. Miller et al., Cancer 1981; 47:207-214). A complete response (CR) was defined as complete disappearance of all evidence of disease for at least 4 weeks. A partial response (PR) was defined as a ≧50% reduction in the product of the two largest perpendicular diameters of the largest measurable lesion, with no new lesions or progression of any lesion, for at least 4 weeks. Progressive disease was defined as a 25%, or more increase in the size of one or more measurable lesions or the appearance of new lesions. Disease not meeting criteria for response or progressive disease was termed stable disease.

Results

A total of 22 patients were enrolled in the study, but five patients were considered inevaluable for efficacy. There were no complete or partial responses. Three patients had disease progression within the first month. Six patients had disease stabilization for more than 3 months (3.5, 3.5, 5, 8, 12+, 14+). Median survival for the entire group was 8 months from the date of diagnosis and 5 months from the start of treatment. One patient with biopsy-proven liver metastases and a CEA of 37 ng/ml (normal <3 ng/ml), had absolute stabilization of the liver metastases and CEA for 8 months. One had stable disease for 5 months. One patient had disease relapse in her pancreatic bed 4 months after a Whipple procedure and was been stable on the composition for at least one year, with the exception of a slowing rising CEA. A third patient had a percutaneous stent inserted and continued to work full-time for at least 14 months with no evidence of tumor progression.

All 22 patients were evaluable for toxicity, having received a total of over 500 injections. None developed any clinical or laboratory evidence of drug-related toxicity. There was no detrimental effect on Quality of Life which generally parallelled disease activity. No significant changes in total white blood cell counts, absolute lymphocyte counts on serum immunoglobulins were seen.

Survival curves representing the survival times from diagnosis and from treatment initiation are presented in FIGS. 13 and 14 , respectively. For comparison, an historical survival curve for Gudjonsson (1987) has been superimposed in FIG. 13 . Another example of a comparable historical survival curve may be found in Bakkevold, Petterson, Arnesjo and Espenhaug (1990).

The results of the survival analyses are summarized in Table 43. The mean survival time for diagnosis was 281 days (FIG. 13 ). The median survival was 182 days (approximately 5 months). For comparison, Gudjonsson (1987) reported the mean survival of his 188 surgical patients as 208 days with a media survival of 120 days. The mean survival time from treatment start was 166 days (FIG. 14 ). The median survival was 133 days (approximately 4 months and 1 week).

TABLE 43
Protocol CO2-104 Survival Estimates
	Patient	Mean Survival	Standard	Median
Survival	Population	(days)	Deviation	Survival (days)
From	Protocol	281	203	182
diagnosis	CO2-104
	patients
	Protocol	304	157	219
	CO2-104
	evaluable
	patients
From	Protocol	166	135	133
treatment	CO2-104
start	patients
	Protocol	220	132	146
	CO2-104
	evaluable
	patients

Survival times were also estimated among a subset of evaluable patients who had each received at least 13 injections. Fourteen of the 22 patient were evaluable. Among these patients (Table 43), the median survival from diagnosis was 219 days (approximately 7 months and 1 week). The median survival from treatment start was 146 days (approximately 5 months).

EXAMPLE 21

Clinical Trials re Malignant Nelanoma

Advanced malignant melanoma was defined to include all stage III or IV patients and all loco-regional or distant relapses occurring after primary treatment. The standard treatment by which all other treatments are judged is DTIC (dacarbazine) which has a reported response rate of about 15%. The median response is 3-6 months, and carries with it severe nausea and vomiting, and a potentially lethal side effect of acute liver necrosis by thrombosis of the hepatic veins. This treatment fails to show any definitive survival advantages.

This study was conducted to determine the safety and efficacy of the composition of the invention and to determine its effect on survival and on quality of life, when used in patients with advanced malignant melanoma. The study, was a non-comparative, multicenter trial.

An initial dosing schedule of 7.5 ml injections of the composition of the invention intramuscularly 3 times. per week was used. After no organ or marrow toxicity was observed, the loading schedule was increased to daily injections for 15 days, followed by maintenance of 3 injections per week. Subsequently the loading dose was increased to 30 days. Duration of treatment was 36 weeks and then reduced to 16 weeks, after which patients were given the option of entering a continuation protocol.

Thirty-three patients with advanced melanoma were included in the study population (17 females and 16 males), ranging in age from 17 to 85 years of age. 64% had been previously treated and 36% were untreated. Of the 33 patients included in the study population, twenty five were evaluable. The Karnofsky Performance Status (baseline) was in the range of 40-100%, median 80%. Eleven patients were alive at the end of the study period and five of these were under treatment.

A minor partial response was observed in 16/33 patients (48%). One patient had a reduction of 33% in the lungs, six patients had pain reductions and eight patients gained more than 1000 grams in weight for more than a month (Range 1000-2600 grams). A stable condition was observed in 19/33 patients (58%) (Range 60-170 days, median 77 days).

FIGS. 15 , 16 and 17 show the survival of patients treated with the composition of the invention compared to historical controls, measured as survival from diagnosis of metastases/recurrence in days. The solid line represents the survival curve for patients treated with the composition of the invention and the broken line represents the historical survival curve (Balch, C. M. et al., Cutaneous Nelanoma, 2nd. ed. 1992, Chps. 14 and 39, pp. 165-187 & 499-508, Lippincott Co., Philadelphia, Pa.). The survival of all patients treated with the composition of the invention, including patients with one to over three tumor sites, is shown in FIG. 15 . Survival of patients with two tumor sites and with three or more tumor sites is shown in FIGS. 16 and 17 respectively.

The group of all patients treated with the composition of the invention had a 39% survival (Kaplan-Meier estimation) at one year. The survival rate at one year for all advanced malignant melanoma (AMM) patients is approximately 11% in historical controls (matched by number of tumor sites). The group had a median survival of 315 days compared to the historical median of 89 days.

With two tumor sites the one year survival was 49% in the patients treated with the composition of the invention, as compared with 13% in historical controls. This group had a median survival of 360 days compared to the historical median of 120 days. With three or more tumor sites the one year survival was 31% in the patients treated with the composition of the invention, as compared with 0% in historical controls. The group with three or more tumors had a median survival of 205 days compared to the historical median of 60 days.

Quality of life was assessed by weight gain, performance status (Karnofsky), Quality of Life Index (Spitzer) and pain scale (Linear Analogue). weight gain over time is shown in Table 39.

TABLE 44
Number of	1st	2nd	3rd	4th	5th	6th
patients/	month	month	month	month	month	month
evaluable	11/25	12/25	4/25	4/25	1/25	1/25
Percent	44%	48%	16%	16%	4%	4%
(%)
Range	100-	200-	100-	100-
(gr)	2400	6000	1000	2000
Average	900	1480	525	775	100	2000
(gr)

The Karnofsky and Spitzer scales are both subjective and were found to approximately agree in each individual. Fifteen patients reported no change in these parameters. Four patients showed fluctuations which later returned to previous levels. One patient had a decrease (from 40-20%).

The results of pain evaluation showed that in six patients by week 4 the pain dropped from 5 (worst possible) to 2 (moderate) or 0 (no pain). One patient had a drop in pain from 3 to 0. One patient with hepatic metastasis had pain reduction to 0 and stabilisation for 11 months. Nine patients who entered the study with 0 pain maintained that level throughout the study. Five patients had a moderate (2 unit) increase in pain. Three patients had transient pain increases (1 to 2 units) during the second or third month.

Out of 1734 injections administered to 33 patients, 21 patients had no adverse drug reactions. Fourteen adverse drug reactions were reported in 12 patients. The adverse drug reactions usually occurred at weeks 4 or 8 and were mild to transient, and most frequently were a low grade fever.

The difference in survival between the historical groups and the protocol groups treated with the composition of the invention suggests a survival benefit for patients treated with the composition of the invention. The cancer seemed to stabilize in 19 patients. All patients treated for ANK were included in the survival data. Also included were 21 previously treated patients (many clinical trials require untreated patients, because of the poor prognosis of failed previous treatments). The tumor burden was high (82% had more than one metastatic site).

The survival and quality of life data suggest that most patients received some benefit from the treatment. Eleven patients were still alive at the end of the study period and of those 11, 5 continued treatment.

EXAMPLE 22

Pathology Protocol Malignant Melanona

The following is a report of a 73 year old female with progressive malignant melanoma of the hard palate and gums. FIG. 25 shows two views of malignant melanoma as seen under the microscope. In FIG. 25 a , looking from top to bottom, one can see the epithelial layer with accompanying keratin, beneath which the malignant cells start to became more apparent. These melanoma cells can be seen to be rounded or oval, with an abundant eosinophilic cytoplasm, and pleomorphic hyperchromatic nuclei. These cells have substituted the normal submucosal tissue. The blood vessels which are seen appear normal, and there is a paucity of any kind of inflammatory/immune response as would be represented by the presence of leukocytes (polymorphonuclear and mononuclear cells). This is an example of tumor tissue which is thriving, i.e. the tumoral architecture is intact.

In FIG. 25 ( b ) a tumor tissue sample is shown from the same patient, who had been treated with the composition for two months. Starting from top to bottom, one can see that the continuity of the epithelium has been disrupted by a necrotic process. This necrosis, while common in the center of any tumor that has reached a critical mass, is rarely seen on the periphery, especially in malignant melanoma, and is a sign that the host's immune response is mounting an attack against the tumor. Throughout the photo are a massive number of cells different from the original tumor cells. These are the immune cells-neutrophils, lymphocytes, macrophages—which have orchestrated the disruption of the typical tumoral architecture. The blood vessel walls have become densely infiltrated with a large number of host immune cells (arrow). This cellular infiltrate subsequently will cause the destruction of the blood vessel, which in turn prevents the tumor from receiving its supply of nutrients and oxygen (ischemic necrosis). This immune response which contributed to the tumoral disruption seen in this patient's tissue slide is consistent with reported changes known to be brought about by TNF (tumor necrosis factor) and with the results of the work described in the previous examples.

The immune response demonstrated in the after treatment with the composition slide ( FIG. 25 b ) strongly links the in vitro TNF immune modulation by the composition with known in vivo anti-tumoral TNF effects.

From the foregoing, it will be appreciated that, although specific embodiments of the invention have been described herein for purposes of illustration, various modifications may be made without deviating from the spirit and scope of the invention. Accordingly, the invention is not limited except as by the appended claims.

EXAMPLE 23

Isolation of Active Fractions

A 300 ml sample of the composition was evaporated to dryness on a rotovap in which the temperature of the bath did not exceed 40° C. In order to ensure that the solution remained basic during the evaporation, 5 drops of a concentrated ammonium hydroxide solution was added every half hour to the composition until the evaporation was complete. The resulting residue had a weight of 11.6 g.

20 ml of a 10% concentrated ammonium hydroxide in methanol solution was then added to 2 g of the above residue. The insoluble material was filtered off and the filtrate was chromatographed through 101.93 g of 60 Å flash silica gel in a column with dimensions of 5 cm×12.5 cm. The solvent system used was 10% concentrated ammonium hydroxide in methanol solution. The column was run at a pressure of 10 p.s.i. and a flow rate of 11 ml/min. After 100 ml of solvent had passed through the column, twelve 20 ml. fractions were collected. The collection of these fractions correlated to the appearance of an off-white band that was quickly moving down the column.

Thin layer chromatography (TLC) of these fractions was run on silica gel plates in a 10% concentrated ammonium hydroxide solution in methanol and visualized with a ninhydrin spray. Fractions having similar TLC profiles were combined, resulting in the following fraction combinations, which were dried on a rotovap.

	Volume Through
	Column to Obtain
Fractions	Fraction	Yield (g)
1-4	100-180	0
5-6	180-220	0.1175
7-8	220-260	0.1969
9-10	260-300	0.0151
11-12	300-340	0.0053
Fractions 5-6, 7-8 and 9-10 had a positive reaction with ninhydrin at an R f value of 0.81.

EXAMPLE 24

Fractions 5-6 and 9-10 from Example 23 were tested in vitro for anti-proliferative effect (in accordance with the procedure of Example 4) and TNF stimulation (in accordance with Example 9). The results are shown below:

FRACTION	ASSAY	ACTIVITY
5-6	Anti-Proliferative Effect	11.57	units/mg
5-6	TNF Stimulation - LPS	50	pg/mg
9-10	Anti-Proliferative Effect	2.6	units/mg
9-10	TNF Stimulation - LPS	1814	pg/mg

Thus, fraction 5-6 was an active anti-proliferative, and fraction 9-10 was an extremely active TNF stimulator, and a moderately active anti-proliferative.

EXAMPLE 25

Samples of Fraction 5-6 was analyzed by Electron Impact Mass Spectroscopy (EI MS) and Electrospray Mass Spectroscopy to identify specific compounds likely to be present in the fraction. The Electrospray MS was performed on a Perkin-Elmer Sciex API-III spectrometer, using 5% acetic acid in water as the solute. In some instances, methanol was added to aid dissolution. The EI MS using a direct insertion probe was performed on a VG Analytical model ZAB-SE spectrometer using glycerol as a matrix, and using a DCI probe on a Kratos Analytical Profile Mass Spectrometer.

A review of the resultant spectra indicated that the following compounds were likely present in Fraction 5-6: phosphocholine, taurocholic acid, choline-stearic acid diglyceride, stearic acid, stearic acid diglyceride, palmitic acid-stearic acid diglyceride, and a sphingosine-oleic acid conjugate.

EXAMPLE 26

100 ml of the composition was acidified with 4 ml of a 1N HCl solution such that the pH of the composition was equal to 3. The composition was then extracted with three 100 ml. portions of HPLC grade dichloromethane. The dichloromethane fractions were then combined and dried over a small amount of anhydrous sodium sulfate. The dichloromethane solution was then filtered through paper into a round bottom flask and evaporated to dryness on a rotovap to yield 0.0049 g of a brown film.

0.0017 g of this film was dissolved in a 214 ppm NH 3 .H 2 O solution at pH 7, and was screened for anti-proliferation activity as is set forth in Example 4. This screen revealed that the solution was an active anti-proliferative, with an activity of 14 Units/mg.

EXAMPLE 27

Example 23 was repeated on a larger scale, as follows. 10 ml of a concentrated ammonium hydroxide solution was added to 900 ml of the composition and the resulting solution evaporated to dryness on a rotovap in which the temperature of the bath did not exceed 40° C. In order to ensure that the solution remained basic during the evaporation, 5 drops of a concentrated ammonium hydroxide solution was added every half hour to the composition until the evaporation was complete, leaving a residue.

150 ml of a 10% concentrated ammonium hydroxide in methanol solution was then added to the total residue. The solution was sonicated for 15 min. and the insoluble material was filtered off. The filtrate was chromatographed through 1695 g of 60 Å flash silica gel in a column with dimensions of 30 cm×12 cm. The solvent system used was 10% concentrated ammonium hydroxide in methanol solution. The column was run at a pressure of 6 p.s.i. and a flow rate of 30 ml./min. The results of the column are summarized in the table below.

	Volume of each
Fraction #	fraction (ml.)	Observations
1	550	colorless
2	450	colorless
3	400	colorless
4	150	colorless
5	100	colorless
6-7	75	colorless
8-13	50	colorless
14	50	tan colored solution begins to elute
15-35	50	tan colored solution
36-40	50	colorless

TLC was run on silica gel plates in a 10% concentration ammonium hydroxide solution and visualized with a ninhydrin spray. Fractions having similar TLC profiles were combined, resulting in the following fraction combinations, which were dried on a rotovap.

	Volume Through
	Column to Obtain
Fraction #	Fraction	Yield (g)	Comments
3	1000-1400	0.0504	white powdery solid
4-5	1400-1650	0.0855	white powdery solid
6-8	1650-1850	0.1555	white powdery solid
9-12	1850-2050	0.3014	white powdery solid
13-14	2050-2150	0.3595	white powdery solid
15-16	2150-2250	0.6914	slight brown color -
			solid is tacky
17-18	2250-2350	1.0284	tan color - solid is
			clumpy
19	2350-2400	0.3432	tan color - solid is
			clumpy
20-23	2400-2600	1.1531	brown color - solid is
			clumpy
24-30	2600-2950	0.8517	brown color - solid is
			clumpy
31-34	2950-3150	0.0813	brown oil

All fraction combinations from 15-16 through Fraction 31-34 had a positive reaction with ninhydrin at an R f value of 0.87, a value very similar to the R f value for the active fractions of Example 23. Fractions 24-30 and 31-34 had an additional positive reaction with ninhydrin at an R f value of 0.85.

EXAMPLE 28

Fractions 4-5, 15-16 and 17-18 were tested in vitro for anti-proliferative effect (in accordance with Example 4) and TNF stimulation (in accordance with Example 9). The results are shown below:

Fraction	Assay	Activity
4-5	Anti-proliferative	4.7 units/mg
	Effect
4-5	TNF Stimulation	—
15-16	Anti-proliferative	4.5 units/mg
	Effect
15-16	TNF Stimulation	—
17-18	Anti-proliferative	3.9 units/mg
	Effect
17-18	TNF Stimulation	—

Thus, fractions 4-5, 15-16, and 17-18 showed anti-proliferative activity, but no TNF stimulation activity. Elemental analysis of the above fractions showed them to be high in NH 4 Cl , which inhibits TNF production.

EXAMPLE 29

Sample of fractions 15-16 and 24-30 was dialyzed and then was analyzed by mass spectroscopy, using the methods described in Example 25. Undialyzed samples from fractions 17-18 and 24-30 were also analyzed. A review of the resultant spectra indicated that the following compounds were likely present: glycocholic acid, a trihexosamine trimer, and taurocholic acid (Fraction 15-16); stearic acid, and a hexosamine dimer; and glycocholic acid (Fraction 24-30).

EXAMPLE 30

The composition was dialyzed in separate dialysis tubing as follows:

100 ml of the composition was placed inside a Spectra/Por™ CE membrane tubing which had a molecular weight cut off of 100. The ends of the tubing were sealed with clips and the tubing was placed into a stirred bath of 10 L of distilled water. The dialysis was monitored daily by removing 1 ml. of solution from the dialysis tubing and adding 3-4 drops of a 1/10 N silver nitrate solution. The presence of chloride indicated that the dialysis was not complete. If the dialysis was not complete the bath was replaced with fresh distilled water. Dialysis completion occurred after 3-4 days. After dialysis was complete, the dialyzed material was dried on a rotovap to yield an average of 0.3 mg of solid per ml of original volume.

A sample of the solid material was then dissolved in HPLC grade water, and TLC was run on silica gel plates in a 10% concentrated ammonium hydroxide solution in methanol, and visualized with a ninhydrin spray. A positive reaction with ninhydrin was obtained at an R f value of 0.83.

EXAMPLE 31

A sample of the solid material from Example 30 was also analyzed by mass spectroscopy, using the methods described in Example 25. A review of the resultant spectra indicated that the following compounds were likely present: a sphingosine-oleic acid conjugate, diacetyl sialic acid, a fucose-hexosamine dimer, deoxyglycocholic acid, taurocholic acid, a sialic acid-fucose dimer, and a di(fucose)hexosamine trimer.

EXAMPLE 32

Previous results indicated that the active components of the composition (at least according to a TNFα release assay) were present in the unbound fractions (void volume) after reversed-phase high performance liquid chromatography (RP-HPLC) on a C18 μBondapack column. As well, most of the mass (70%) of the composition extract, which was loaded onto a C18 μBondapack column, eluted in the void volume. These results suggested the active components of the composition are, very likely, very polar or even ionic molecules.

To further examine the above results, purification of the active components by ion-exchange chromatography was performed. Negatively charged active components (assessed by its anti-proliferation affect on tumor cells and not on normal cells, and as wall by its TNFα release-inducing activity), if present, would thus become bound to an anion exchange resin.

Experimental Procedure

10 ml of total virulizin extract was loaded onto an anion-exchange chromatography column (Bio-Rad AG-1, hydroxide form, total resin wet volume was 10 ml (column dimensions 1.5 cm×6.0 cm), equilibrated with Millipore deionized water). The volume of resin used was calculated to be sufficient for the binding of all the anions present in the extract. The unbound fraction was collected and reloaded onto the column in order to maximize the binding to the resin. The unbound fraction from this second passage was collected and saved. Any unbound material remaining on the column's void volume was removed by washing with deionized water (2×20 ml). Bound molecules were eluted with a step gradient of ammonium bicarbonate (NH 4 HCO 3 ) (20 ml/step).

The elution steps were 0 M
                       0.1 M
                       0.2 M
                       0.3 M
                       0.4 M
                       0.5 M
                       0.6 M
                       1.0 M
                       1.5 M

EXAMPLE 33

Samples from all the fractions of Example 32 were analyzed for anti-proliferation activity and TNF stimulation activity, in accordance with the procedures of Examples 4 and 9, respectively.

The results are shown below:

	Anti-proliferative	TNFα release-inducing
Sample	activity (U/mg)	activity-LPS (pg/ml)
0 M	0	35
0.1 M	0	−79
0.2 M	0	−76
0.3 M	0
0.4 M	2.5
0.5 M	555.6
0.6 M	0	107
1.0 M	0	105
1.5 M	0	189

The results from the activity assays show that TNF production stimulation was found in the 0.6 M, 1.0 M, 1.5 M fractions. Anti-proliferative activity was found in fractions 0.4 M (minor activity) and 0.5 M (major active fraction).

EXAMPLE 34

Thin layer chromatography analysis of the active fractions revealed a mixture of several components. A sample of the 1.0 M fraction from Example 32 was analyzed by mass spectroscopy in accordance with Example 25. A review of the spectra generated suggested that the following compounds may be present: a sialic acid-glycerol dimer, cholesterol sulfate, and taurocholic acid.

EXAMPLE 35

Reversed Phase (C18) HPLC analysis was performed on a sample of the composition in accordance with the procedure of Example 2, except that (1) a Phenomenex WP60009-C18 column, 250×4.6 mm, was used, (2) the sample was lyophilized and then reconstituted in 0.1% trifluoroacetic acid (TFA) in water, and (3) 150 μl of the reconstituted sample was applied to the column.

Various fractions of eluent were collected, including a fraction which eluted at approximately 2.40-3.40 minutes after the reconstituted sample was applied to the column.

EXAMPLE 36

A sample of the fraction eluting at approximately 2.40-3.40 minutes from Example 35 was analyzed three times for TNF stimulation activity in accordance with the procedure of Example 9. The following results were obtained.

Assay # TNF Stimulation - LPS
1       259 pg/ml ± 107 pg/ml
2       311 pg/ml ± 14 pg/ml
3       572 pg/ml ± 176 pg/ml

EXAMPLE 37

A sample of the fraction eluting at approximately 2.40-3.40 minutes from Example 35 was subjected to tandem column reversed-phase (C18) HPLC as follows. The column from Example 35 was used in tandem with a Phenomenex prime-sphere HC-C18 column, 250×4.6 mm. The sample was lyophilized, reconstituted in 0.1% TFA in water (Buffer A) and 150 μl of reconstituted sample was applied to the column. Buffer A was run for twenty minutes, then a linear gradient of 0-80% of 0.1% TFA in acetonitrile (Buffer B) was run for 35 minutes. At the end of this period, 80-0% Buffer B was run for 5 minutes. Flow rate was 0.9 ml/min. Six eluent fractions were collected, at the following approximate times from injection:

Fraction # Time (min.)
1          5.6-6.25
2          6.25-6.6
3          6.6-7.1
4          7.1-8.2
5          8.8-9.6
6          14.7-16

EXAMPLE 38

A sample of Fraction 1 (“1”) and a sample of Fraction 2 (“2”) from Example 37 were lyophilized and reconstituted in 214 ppm NH 3 .H 2 O. These reconstituted samples were then analyzed for anti-proliferative effect in accordance with the procedure of Example 4. The following results were obtained:

Sample Anti-proliferative Effect
1      17.8 unit/ml
2      0

EXAMPLE 39

Samples from each of the six fractions of Example 37 were analyzed by mass spectroscopy in accordance with Example 25. A review of the resultant spectra for the six fractions indicated that the following compounds were likely present: taurocholic acid, a sialic acid-glycerol dimer, NaCl, trimethylamine, methylethylamine, and propylamine.

EXAMPLE 40

Anti-proliferative effect, according to the method of Example 4, was measured for the following three samples: (1) 10-11 mg taurocholic acid in 2 ml of H 2 O; (2) 214 ppm NH 3 .H 2 O; and (3) 0.7 mg of taurocholic acid in 4.0 ml of 214 ppm NH 3 .H 2 O. Neither sample (1) nor sample (2) had any detectable anti-proliferative effect. Sample (3), however, had an anti-proliferative effect of 14 units/mg.

This result indicates that bile acids, such as taurocholic acid, in combination with ammonium ions, exhibits anti-proliferative activity at concentrations below that which the components, tested individually, show no activity. Thus, the combination of these two components, apparently synergistically, affects anti-proliferative activity.

Claims

1. A process for preparing a composition for use as an immunomodulator comprising small molecular weight components of less than 3000 daltons, and having the following properties:a) is extracted from bile of animals; b) is capable of stimulating monocytes and macrophages in vitro; c) is capable of modulating tumor necrosis factor production; d) contains no measurable level of IL-1, IL-1b, TNF, IL-6, IL-8, IL-4, GM-CSF or IFN-gamma; e) shows no cytotoxicity to human peripheral blood mononuclear cells; and f) is not an endotoxin, said process comprising (a) mixing bile from an animal with an equal volume of an alcohol to produce a bile/alcohol solution; (b) separating out the alcohol soluble fraction and isolating a solution substantially free of alcohol; (c) removing bile pigments from the solution to obtain a colorless liquid; (d) treating the colorless liquid to substantially remove any residual alcohol; (e) extracting the colorless liquid with ether and isolating the aqueous phase; and (f) removing residual ether from the aqueous phase.

2. A process as claimed in claim 1 wherein prior to step (e) the colorless liquid is concentrated to about one eighth of the volume of the bile/alcohol solution and after step (f) the aqueous phase is concentrated so that it is one tenth of the volume of the bile/alcohol solution.

3. A process for preparing an immunomodulator composition comprising (a) mixing bile from an animal with a water-soluble solvent to produce a bile/solvent solution; (b) isolating an aqueous solution substantially free of solvent from the bile/solvent solution; and (c) removing bile pigments from the substantially solvent-free solution to obtain a colorless liquid.

4. The process of claim 3 wherein the water soluble solvent is an alcohol.

5. The process of claim 4 wherein the bile from an animal is mixed with an equal volume of an alcohol.

6. The process of claim 3, further comprising concentrating the colorless liquid to about one-eighth the original volume of the bile/solvent solution.

7. The process of claim 3, further comprising concentrating the colorless liquid to about one-tenth the original volume of the bile/solvent solution.