A pharmaceutical product having an anti-tumor activity; the use of a pharmaceutical product or of pharmaceutical compositions in an anti-tumor therapy.

The invention relates to a pharmaceutical product having anti-tumor activity, and to the use of a pharmaceutical product or of pharmaceutical compositions in an anti-tumor therapy .

It is known that endotoxins - toxic materials which are integral components of the cell wall of gram-negative bacteria, are mainly composed of lipid and polysaccharide and hence are sonetimes called lipopolysaccharides ⁽ PS ⁾ or glycolipids - possess many types of anti-tumor activity, including the induction of hemorrhagic necrosis and the regression of tumors in animals. For clinical application, however, they are unsuitable on account of their high toxicity.

In the ^' mean time, detoxified endotoxins, such as non- toxic lipid A, have been obtained for example by acid » hydrolysis with hydrochloric acid (Westphal and Lϋderitz,

60 Angew. Chem. 66, 407, 1954) or irradiation with 150 kGy Co gamma radiation (Fust et al, Infect. I mun . 16, 26-31, 1977) , and synthetic, non-toxic analogues thereof (Kotani et al, Infect. Immun. 41, 758-773, 1983; Kiso et al, Agric . Biol. Chem. 45, 1523-1525, 1981; Behling et al , J. Immu¬ nology 117, 847, 1976; Ponpipom et al , J. Med. Chem. 23, 1184-1188, 1980) , which compare favourably with the endo¬ toxins in a considerably lower toxicity. Unfortunately, this reduction of toxicity is accompanied by a considerable decrease in anti-tumor activity.

Ribi et al (Cancer Immunol. Immunother. 7, 43-58, 1979; Cancer Research 41, 2654-2657, 1981; Cancer Immunol. Immunothe 12, 91-96, 1982) have found that detoxified endotoxin compositions on intralesional administration (i.e. administra¬ tion within the tumor) exhibit a strong anti-tumor activity in guinea pigs with a subcutaneously transplanted line-10

hepatocellular carcinoma induced by a carcinogen, provided the detoxified endotoxin compositions were combined with trehalos≥ dimycolate, muramyl dipeptide, or a bacterial peptide fraction, and oil. It was also found that the addition of muramyl dipeptide dramatically increases the lethality of endotoxins, but that the combination of muramyl dipeptide with detoxified endo¬ toxin does not cause any toxicity phenomena, such as death and lethargy of the test animals.

The need of co-administering oil and of intralesional application, however, render it impossible to transfer the results of research obtained by Ribi et al in guinea pigs direct to medicine. The muramyl dipeptide used by Ribi et al (MDP, chemical name: N-acetylmuramy1-L- alanyl-D-isoglutamine) is a well-known immuno-adj uvant for the formation of specific antibodies against soluble and particulate antigens (Chedid et al, Progr. Allergy, 25, 63-105, 1978) , which is known for example upon the addition of mineral oil, e.g. Freund's incomplete adjuvant, to become a good adjuvant for specific, cell- mediated immune reactions (Chedid et al, J. Reticulo— endothel. Soc. 26, 631-640, 1979) . MDP is not immunogenic itself. Also, MDP is capable of stimulating various non¬ specific effector mechanisms of the immune system, including resistance to pathogens and the production of interleukins (Parant, Springer Semin. Immunopathol . 2, 101-118, 1979; Wood et al, Cell. Immunol. 70, 407, 1982) . MDP does not by itself, however, exhibit any significant anti-tumor activity.

In summary, the following properties of MDP are known: A. stimulation of the formation of specific antibodies

(humoral adjuvant) B stimulation of specific cellular immunity (cellular adj uvant)

C. stimulation of non-specific resistance

D. interferon induction (lymphocytes and/or monocyte acro- phage factor)

E. interleukins induction (leukocyte factors)

F. macrophage activation

G. undesirable effects, such as pyrogenicity (fever-inducing capacity) , induction of temporary leukopenia (shortage of circulating white blood cells) , induction of inflammat of joints (arthritis) and increased sensitivity to the ^• toxicity of endotoxins. Some of these properties are coupled to the use of oil. Various non-pyrogenic analogues and prodrug forms of MDP have been described (Kiso et al, Carbohydrate Research 90, C8-C11, 1981; Lederer, Prog. Immunol. IV, 1194-1211, 1980; Krueger et al, J. Exp. Med. 159, 68-76, 1984; Durette et al, Carbohydrate Research 108, 139-147, 1982; Matsumoto et al, Infection and Immunity 39, 1029-1040, 1983; Ogawa et al, Infection and Immunity 35, 612-619, 1982; Saiki et al, Infection and Immunity 39, 137-141, 1983; Chedid et al, Proc. Natl. Acad. Sci. USA 73, 2472-2475; Lederer, J. Med. Chem. 23, 819, 1980; Matsumoto et al, Infection and Immunity 32, 748-758, 1981; Tsujimoto et al , Micro- biol. Immunol. 23, 933-936, 1979; Durette et al, J. Med. Chem. 25, 1028-1033, 1982; Lefrancier et al, J. Med. Chem. 25, 87-90, 1982; Masihi et al, Infection and Immunity 43, 233-237, 1984; Kusu oto et al, Bull. Chem. Soc. Japan 52, 1665-1671, 1979; Arnon et al, Proc. Natl. Acad. Sci. USA 77, 6769-6772, 1980) . The properties of all these known analogues and prodrug forms exhibit quite some variety. Thus, compared with MDP itself, the analogue 6-0- (2-tetradecyl-h xadecanoyl) -N-acetylmuramy1-L-alany1- D-isoglutamine is a stronger stimulator of cellular immu¬ nity and a weaker stimulator of non-specific resistance; the analogue 6-0-acetyl-N-acetylmuramy1-L-alany1-D- isoglutamine is as active as a stimulator of non-specific resistance and equivalent to MDP as adjuvant, and the analogue N - (N-acetylmu amyl-L-alany1-D-isoglu- tamyl)-N -stearoyllysine is a stronger stimulator of non¬ specific resistance than MDP.

It has now surprisingly been found that the use of a combination of detoxified endotoxin (or a synthetic, non-toxic or acceptably toxic analogue thereof) and mu¬ ramyl dipeptide (or a non-pyrogenic or acceptably pyro¬ genic analogue or prodrug form thereof) , without the oil and trehalose dimycolate that is necesgary according to Ribi et al, upon administration other than intralesional,

in particular upon intravenous, intramuscular and/or subcutaneous administration, leads to a synergistic effect as regards anti-tumor activity, without being accompanied by a substantial increase in toxicity.

The invention accordingly relates to a pharmaceutical product having an anti-tumor activity, comprising one or more pharmacologically acceptable vehicles, solvents, diluents and/or adjuvants, and (a) at least one detoxified endotoxin or a synthetic, non-toxic or acceptably toxic analogue thereof, and (b. muramyl dipeptide (I1DP) , or a non- pyrogenic or acceptably pyrogenic analogue or prodrug form thereof in quantities which on the one hand are toxicologically acceptable and, on the other hand, posses an effective anti-tumor activity.

In one embodiment of the invention, the pharmaceutical product has the form of one single solution for intrave¬ nous, intramuscular or subcutaneous administration. In a preferredembodiment of the invention, however, the pharmaceutical product has the form of a solution for intravenous, intramuscular or subcutaneous administration of at least one detoxified endotoxin or a synthetic non- toxic or acceptably toxic analogue thereof, and a separate solution for the intravenous, intramuscular or subcutaneous adminis ration of MDP or a non-pyrogenic or acceptably pyrogenic analogue or prodrug form thereof. By separate administration, i.e. administration at different locations, undue formation of antibodies against component (a) , resulting from the immuno-adjuvant activity of (b) , and hence a neutralisation of the effect upon a subsequent injection or possible anaphylactic reactions against (a) can be reduced.

The invention further relates to the use of the pharmaceutical product according to the invention in an anti-tumor therapy.

More specifically, the invention relates to the use of a pharmaceutical composition comprising at least one detoxified endotoxin or a synthetic, non-toxic or accept¬ ably toxic analogue thereof, in combination with

the use of a pharmaceutical composition comprising MDP or a non-pyrogenic or acceptably pyrogenic analogue or prodrug form thereof, in an anti-tumor therapy in which the two compositions are administered simultaneously or virtually simultaneously at different sites in quantities which on the one hand are toxicologically acceptable and, on the other hand, have an effective anti-tumor activity

Preferably the two compositions referred to are solu¬ tions for intravenous, intramuscular and/or subcutaneous administration, in particular aqueous solutions. Other ways of administration, for example, intraperitoneal administration, however, are possible.

The invention is illustrated in and by the following examples .

EXAMPLE 1

In this example, the effect was investigated of intra¬ venously (iv) injected aqueous solutions of MDP in combi¬ nation with a detoxified endotoxin composition against solid subcutaneous (sc) Meth A fibrosarcoma transplants in syngeneic BALB/c mice. For comparison, the effect of i.v. injections of aqueous solutions of MDP alone, detoxi¬ fied endotoxin alone, MDP+endotoxin , and of a saline solution alone were also investigated.

For the experiments, 11 weeks old female BALB/c mice were used. The Meth A sarcoma of BALB/c origin was ob¬ tained from the Clinical Research Centre (Harrow, Middle¬ sex, Great Britain) and maintained in the ascites form by serial intraperitoneal passage.

The MDP was obtained from the Institut Pasteur Produc¬ tion (Marnes-la-Coquette , France) . Endotoxin (LPSw) of E_. coli 0111-B4 (Difco Laboratories, Detroit, Michigan, USA) was prepared by the phenol/water extraction method (Westphal et al, Angew. Chem. 66, 407-417, 1954 and Z. Naturforsch. 76, 148-155, 1952) .

Purified toxic and detoxified endotoxin of the heptose- less (Re) mutant Salmonella typhimurium (Ribi Immuno Chem. Inc. , Hamilton, Montana, USA) were prepared and purified as described by Ribi et al, Cancer Immunother. 12, 91-96, 1982, which in brief came down to the extraction

of endotoxic glycolipids from cell walls by means of phenol/ chloroform/petroleum ether, fractionation by means of preparative microparticu__sfce gel chromatography and further purification on Sephadex LH-20. In this way, disaggregated endotoxic glycolipids were obtained that were free from phospholipids and nucleic acid and had a protein content of less than 0.3%. The detoxification of the purified endotoxin was carried out by hydrolysis at 100°C in 0.1 N HCl. The resulting composition was 1000 times less toxic as determined by the lethality for chicken embryos, was practic- ly free from KDO (keto-3-deoxyoctonate) and possessed approximately half the phosphate content of the purified endotoxin.

The MDP was dissolved in pyrogen-free saline. The endotoxins were dissolved in o.5% triethylamine (2 mg endotoxin/0.4 ml) and diluted in saline. A total volume of 0.5 ml was injected.

The mice received a s.c. injection with 3 x 10 viable

Meth A cells in the abdomen and were treated 9 days laters,wh the tumor had a diameter of approximately 7.5 mm. Necrosis was measured on the 11th day and expressed as 100 times the ratio of mean diameters of necrotic area and tumor. The incidence of dark and light stained necrosis was separately indicated. Growth inhibition was scored when the tumor had not increased in size within 2 days after the injection. Complete cure was scored when the tumors disappeared com¬ pletely within 19 days of treatment. The data are expressed as the mean value ± SEM = standard error of the mean) .

The results are summarized in the following tables A and B .

TABLE A

Anti-tumor activity of E. coli endotoxin and/or MD

incidence ( % ) treatment necros is of

Endotoxin. MDP number incidence (%) growth com-

J3. coli 0111 :B4 dose of extent* inhibi- plete dose (μg) (ug) mice light dark tion cure

25 0 10 30 70 57 + 5 80 30

10 100 10 0 100 75 + 4 100 90 10 30 20 0 100 68 + 2 100 80 10 10 5 0 100 66 + 3 100 100 10 3 5 0 100 57 + 3 100 80 10 0 20 50 35 46 + 3 65 15

3 30 10 0 100 61 + 4 100 60 3 10 10 0 90 66 + 2 100 60 3 0 14 50 7 49 + 3 43 7 1 30 5 0 100 63 + 5 100 80 1 0 5 40 0 36 + 1 0 0 0,3 30 5 20 80 56 + 7 100 60

0 L00 10 0 0 30 20 0 30 15 0 0 7 7 0 10 4 0 0 25 25 0 3 5 0 0 0 0 0 0 20 0 0 5 0

* mean ± SEM

Table B

Anti-tumor activity of toxic and detoxified endotoxin of S. typhimurium Re alone with MDP incidence (%] treatment necrosis of n incidence growth com¬ endotoxin dose MDP inhibi¬ plete

S_. typhimurium dose tion cure iRe (μg) (yg) extent light dark

50 detoxified 0 5 20 0 31 0 0

25 ^■ 0 10 60 0 36 ± 6 10 0

10 30 10 10 90 61 + 3 80 70

10 0 10 10 0 47 40 10

toxic 10 30 5* 20 80 73 ± 4 100 80

10 0 - 10 ^" 50 50 57 ± 5 90 40

saline 0 30 10 0 0 0 0 0 0 10 0 0 0 0

of ten mice treated 5 died within 24 h

Furthermore, the toxicity of i.v. injected toxic and detoxified S_. typhimurium Re endotoxin, alone or mixed with varous MDP doses, was determined. Results are shown in the accompanying figure. Groups of 5 mice received laboratory food and water ad libitum, and were kept in an atmosphere of constant temperature (18°C) and humidity. The change in body weight, measured 24 hours (open dots) and 48 hours (black dots) after in¬ jection were expressed proportionally to the individual weight, as these were immediately preceding the injection and shown as mean ± SEM (vertical bars) . The figures in brackets indicate the total number of dead mice after 24 and 28 hours. Diarrhea and lethargy were evaluated after 5 hours. Ratings indicate the number of mice having these symptoms. The designation nd means: not determined.

In the experiments, the results of which are shown in Table A, it turned out that the injection of 25 μg __.• ^co -*-**- endotoxin in BALB/c mice having 9 days old Meth A tumors caused in 100% of the cases a marked tumor necrosis, followed by growth inhibition of most tumors and complete cure in 30% of the mice. Lower doses were less active in all respects. MDP alone caused. in various doses, no tumor necrosis, but did induce some regression. The addition of 3-100 μg MDP to 10 ug E_. coli endotoxin led to a considerable enhancement of all measured parameters of anti-tumor activity of the endotoxin. with the exception of the extent of necrosis, no clear relationship between dose and effect was observed. A decrease in endotoxin dose with a constant MDP dose of 30 μg did not lead to a clear decrease in anti-tumor activity.

In the experiments, the results of which are shown in Table B and the figure, it turned out that detoxified S_. typhimurium Re endotoxin in various doses caused very slight necrosis, and hardly affected tumor growth.

The toxic endotoxin exhibited a considerable activity in a dose as low as 10 μg, but the addition of MDP was often lethal. The combination of the detoxified endo¬ toxin composition with MDP, however, led to very strong anti-tumor activity, while none of the combinations in¬ vestigated caused lethality or diarrhea, and lethargy was observed with higher MDP doses only. The loss in weight increased with increasing doses of both components, but never exceeded 10%. Body weight after 48 hours was in all cases higher than after 24 hours, which appears to indicate recovery. As shown in the figure, all doses of toxic endotoxin, whether or not combined with MDP, caused lethargy and diarrhea, although MDP surprisingly turned out to offer some protection against diarrhea. Weight loss was substantial, sometimes more than 15%, and after 48 hours greater than after 24 hours. Higher MDP doses led to decreased loss in weight, which could be indicative of fluid accumulation in tissues as a result of shock. At higher doses, lethality was observed. The LD _ςn of toxic endotoxin alone is more than 160 ''g per mouse, of a com¬ bination with 30 ^' g and 100 ^r g MDP however, 28.3 .g and 24.6 g, respectively (calculated by the Spearman-Karber method modified by Nowotny, Basic Exercises in Immuno- chemistry, 303-304, 1979, Springer Verlag, Berlin) .

EXAMPLE II

In this example, the effect was investigated of the time and the route of administration of the endotoxin and MDP. For the animal experiments, the same kinds of mice and tumors were used as in the preceding example.

In addition to endotoxin of E. coli 0111 :B4, endotoxin

(LPSw) of E. coli 089 and a radio-detoxified (150 kGy,

60

Co gamma irradiation) preparation of the same endo¬ toxin were used, both obtained from prof. dr. L. Bertόk of the "Frederic Joliot-Curie" National Research Insti¬ tute for Radiobiology and Radiohygiene, Budapest, Hungary.

•_x

_? 4?

^■

The detoxification procedure has been described by Fust et al, Infect. Immun. 16, 26-31, 1977; the detoxified preparation is commercially available under the trade- mark of Tolerin . The LD values of these two preparations in female CFY-random rats were 2 mg/kg and 21 mg/kg, res¬ pectively. Unless otherwise indicated, the preparations were i.v. injected in 0.5 ml pyrogen-free saline.

The mice received a subcutaneous injection with 3 x 10 viable Meth A cells in the abdomen on day 0. Treatment was started on day 9 (when the tumor had a diameter of approximately 7.5 mm) unless otherwise indicated. The extent of light and dark stained necrosis was measured on day 11 and expressed as 100 times the ratio of mean diameters of necrotic area and tumor. Diameters were measured with a caliper. Growth inhibition was scored if the tumor had not increased within two days of treat¬ ment. Complete cure was scored if the tumors had disappear¬ ed completely wthin 19 days of treatment. Data are expressed as the mean value ± SEM. Analysis for significance was performed by Student's t test. P values over 0.05 were considered non-significant.

The results are summarized in the following tables C-F.

TABLE C

Effect of variation of the lime of MDP administration on the aniiiumor activity of endotoxin (£. coli 0111 : B 4)

Expt no. Treatment* n Nccroju ^1* Incidence (9, -)of

EndotoxinMDP Incidence (%) Extent Gto ih Complete inhibition* cure ¹

Light Dark

1 + — 10 60 30 S0±3 60 20

- + (Oh) 10 10 0 43 10 10

*► + (-4βh) 10 50 30 41 ±3 20 10

+• + (-2 h) 10 BO 0 49 ±4 80 30

+ + (Oh) 10 0 100 66±3 ^C 90 60

^• + (+24 ) 10 70 30 44±3 SO 0

- - 10 0 0 0 0

2 + - 5 40 40 51 ±4 60 20

- + (Oh) 5 20 0 47 20 20

+ + <-4h) 5 20 80 68 ±3* 100 80

. + + (Oh) S 0 100 63 ±6 80 80

+ + (+ ) 5 0 100 57 ±7 100 100

- — 5 0 0 0 0

B JIc mice with -^y*-old Meth ΛlranφlanU-Ccnved 3 μgendotoxw^ saline waa injected IV before, together with (Oh), or after endotoxin

Ligbt and dark necrosis were measured oα day 11. Extent is given as mean ± SEM

F < 0.05 compared with all other treatments in expt 1

F < 0.05 compared with treatment with endotoxin alone

No increaac of tumor so* within 2 days after injection

Complete disappearance of the tumor within 19 days of treatment

TABLE D

Anti-tumor activity of MDP and endotoxin (E_. coli 0111:B4) , injected simultaneously by different routes.

treatment ^' necrosis incidence (%) of footpad i.v. incidence (%) ext. ent. g _in ro _h w _i _th _p co _l m _et -

Light dark bition ^J cure' s a l ine endotoxin 90 0 41 + 5 40 30

MDP s aline 0 0

MDP endotoxin 10 90 62 ± 4 ^" 70 50 endotoxin s a lin e 0 0 0 0

endotoxin MDP 0 100 59 ± 4 ^' 70 20 s aline s a line 10 0 0 0

Groups of 10 BALB/c mice with 9 days old s.c. Meth A tumors on the abdomen were injected simultaneously with

3 ^' g endotoxin and 30 g MDP s.c. in the footpad and i.v. , respectively, or the other way round. 2 Necrosis was measured on day 11. Extent is given as the mean value ± SEM.

No increase of tumor size within 2 days after injection. 4 Complete disappearance of tumor within 19 days of treat¬ ment . p < 0.05 compared with corresponding treatment without MDP .

^•

TABLE E

Anti-tumor activity of E_. coli 0111.B4 endotoxin and MDP injected s.c. at different sites.

tr 4- 1 . 2 eatment necrosis incidence (%) of inciden ce footpad footpad growth complete (%) left right extent inhi- -. cure bition light dark endotoxin saline 0 0 0 0

endotoxin MDP 80 0 42 ± 4 40 60

Groups of 5 BALB/c mice with 9 days old Meth A transplants were injected s.c. with 3 g endotoxin (LPSw of E_. coli

0111.B4) , and 30 g MDP or saline in the left and right footpad, respectively. 2 Measured on day 11. Extent is expressed as mean va'lue

± SEM.

No increase of tumor size within 2 days after injection. 4 Complete disappearance of tumor within 19 days of treat¬ ment .

TABLE F

Anti-tumor effect of radio-detoxified (Tolerin ) and normal endotoxin of E. coli 089 alone or in combination with MDP.

i.v. treatment 2 necrosis incidence (%) of n frequency (%)

Endotoxin Dose MDP extent growth CO m lete light dark inhi- 4 cure bition

Tolerin 25 ^' μg — 5 60 0 48 + 6 40 0

Tolerin 10 μg - 10 40 10 36 ± 3 30 0

Tolerin 10 μg + 10 0 100 59 + 3 60 20

E. coli 089 25 μg - 5 20 60 50 + 5 t 40 20

E. coli 089 10 μg - 10 70 20 43 ± 4 30 0

E. coli 089 10 μg + 10 0 100 61 ± 3 100 50 saline + 10 0 0 0 0

saline - 10 0 0 0 0

BALB/c mice with 9 days old Meth A transplants were injected i.v, with an endotoxin .preparation alone or mixed with 30 g MDP in 0.5 ml saline. ^' Scored at day 11. Extent is given as mean value ± SEM.

No increase of tumor size within 2 days after injection. Complete disappearance of tumor within 19 days of treatment p < 0,05 compared with treatment without MDP.

The experiments showed that the mere i.v. administration of a suboptimal dose of 3 μg E_. coli 0111.B4 endotoxin to Meth A bearing mice caused a high incidence of predominantly light colored necrosis, followed by tumor growth inhibition in 60% ^? of the mice and complete cure in 20% of the mice. MDP alone caused only incidental necrosis and regression. Mixed administration induced 100% incidence of dark- colored and significantly more extensive necrosis and a high incidence of complete regression (60%) as compared to treatments with either components alone. The administrati of MDP one or two days before or one day after endo¬ toxin did not lead to an enhancement of anti-tumor activi¬ ties of endotoxin. But administration of MDP and endoto::ir within 4 hours of each other lead to the same ef.ect as simultaneous administration.

The simultaneous administration of endotoxin and _^ -MDP by the i.v. and s.c. route, respectively, caused a com¬ parable anti-tumor activity to that caused by combined i.v. administration. Endotoxin injected by the s.c. route lacked any anti-tumor activity. When MDP was simultaneous.- ly injected i.v. however, an equally strong necrosis and tumor growth inhibition were achieved as in the case of a combined treatment by the reversed route, but this was less frequently followed by regression. The simultaneous injection of endotoxin and MDP by the s.c. route in the left and right footpad, respectively, caused both necrosis and a high incidence of regression, while here again endo¬ toxin alone was ineffective.

Table F shows that a dose of 25 llq __. coli 089 endo¬ toxin induced a reasonable degree of necrosis and 20% regressions. Tolerin was less active in the same dose. Although smaller quantities (10 _Ag) of both substances possessed a lower activity, anti-tumor activity was potentiated by the addition of MDP to these low doses of endotoxin preparations to a higher value than was observed upon the administration of 25 Λg of either endotoxin preparation alone.

Examp le I I I

This example makes a further analysis of the po¬ tentiating action of MDP by use of several analogues of MDP known to differ in pyrogenicit , adjuvant action and capa¬ city to stimulate nonspecific resistance. Two other synthetic agents with adjuvant action in aqueous vehicles, ^' (Snippe et a Int .Arch. llergy Appl . Immunol . 65, 390, 1981) the pluronic polyol L121, a good adjuvant for the induction of antibody formation and dimethyldioctadecylammonium bromide (DDA) , an adjuvant which also favours the induction of cell-mediated immunity, were compared with MDP and MDP analogues for their activity .

As in example 1, female BALB/c mice were used at the age of about 11 weeks. The syngeneic and immunogenic Meth A fibrosarcoma was obtained from the Clinical Research Centre (Harrow, Middlesex, Great Britain) and maintained in the as- cites form by serial IP passage.

Endotoxin (LPSw from E. coli 0111.B4) was bought from Difco Laboratories (Detroit, Michigan, USA) ; dimethyldioctade cylammonium bromide (DDA) from Eastman Kodak (Rochester, New York, USA) ; pluronic polyol L121 from BASF Wyandotte Co, ( yandotte, Michigan, USA) ; MDP from the Institut Pasteur Production (Marnes-La-Coquette , France; designated as MDP(F)) N-acetylmuramyl-L-alanyl-L-isoglutamine (MDP(L-L)) from Cal- biochem-Behring (La Jolla, California, USA) . MDP, designated as MDP(J) and N - (N-acetylmuramyl L-alanyl-D-isoglutaminyl) - N -stearoyl-L-lysine (MDP-Lys (Ll8 ) ) were kindly provided by drs Yokoyama and A. Inoue (Daiichi Pharmceutical Co. , Tokyo, Japan) ; 6-0- (2-tetradecylhexadecanoyl-N-acetylmuramy1-L-alan D-isoglutamine (B30-MDP) and 6-0-acetyl-N-acetylmuramyl-L- alanyl-D-isoglutamine (L2-MDP) were generous gifts from profs S.Kotani and T.Shiba (Osaka University Dental School and Osak University, Osaka, Japan) . All agents but MDP-Lys (L18) and B30-MDP were dissolved in pyrogen-free saline. The latter agents were usually dissolved in 0.03 M phosphate buffer (pH 7.5) with 5% (w/v) Nikkol HCO-60 , a nonionic detergent for clinical use (Nikoh Chemicals, Tokyo, Japan) and 4% (w/v)

glucose. If necessary further dilutions were made with the Nikkol solution or saline. In one experiment MDP-Lys (L18) was administered in saline after sonication for 5 sec. at maximal output with a Mullard Sonifier type 7685/2. The same was done for DDA and L121 Stock solutions of endotoxin and MDP in saline were stored in small aliquots at -20 °C. Unless other¬ wise indicated solutions of MDP analogues DDA and L121 were prepared just prior to use.

Mice received a SC injection with 3 x 10 viable Meth A cells in the abdomen and were treated IV with one or more agents in 0.5 ml vehicle/saline 9 days later (tumour diameter ± 7.5 mm) . The extent of light and dark stained necrosis was measured with a caliper at day 11 and expressed as 100 times the ratio of the mean diameters of necrotic area and tumour. Duration of temporal regression was represented by the number of days that the tumour size was smaller than or equal to the tumour size at the day of treatment. Complete regression was scored when tumours disappeared within 19 days of treatment.

Data have been expressed as mean ± SEM. Analysis for significance was done by Student's t_ test. P-values over 0.05 were considered not significant.

The results are summarized in the following tables G-K.

Table G. Potentiation of endotoxin-induced tumour necrosis and regressio by MDP and various analogues.

Exp. Treatment n Necrosis Incidence (%) of regression no.

Endo- MDP Incidence (%) Ex- Tem¬

(Duration) Comp toxin analogue tent poral

Light Dark

+ Saline 5 60 20 46±6 60 (> 8±6) 20

+ MDP(P) 5 0 100 69±2 ^C 40 (>12±7) 60

+ MDP(J) 4 0 100 69±3° 50 ( 4±1) 50

+ L2-MDP 5 0 100 66±8° 60 (>13±6) 40

+ MDP-Lys (L18) 5 0 100 74±4 ^C 80 ( 11±3) 20

+ B30-MDP 5 0 100 71±3 ^C - 100

- MDP(F) 4 0 0 0 25

- MDP(J) 4 25 0 31 0 0

- L2-MDP 4 0 0 0 25

- MDP-Lys (L18) 4 0 0 0 0

- B30-MDP 4 25 . ^' o 32 25 ( 15 ) 0

— Saline 5 0 0 0 0

BALB/c mice with 9 day old Meth A sarcoma transplants were injected IV with 3 μg endotoxin alone or combined with 60 nmol of an MDP analogue (60 nmol MDP -v 30 μg) in a total volume of 0.5 ml saline, except MDP- Lys (L18) and B30-MDP which were injected in 0.5 ml 3% Nikkol in saline.

Light and dark stained necrosis were measured on day 11. Extent is given as mean ± SEM.

p < 0.05 compared to treatment with endotoxin alone.

Complete regression was scored when the tumour had disappeared within 19 days of treatment. Temporal regression was scored when tumour size after treatment was _. tumour size at the day of treatment during one or more days. Duration is given as mean number of days ± SEM.

Table H. Potentiation of endotoxin-induced tumour necrosis and regression by different doses of MDP and various analogues.

a n . b Ex - Treatment Necros s Incidence (%) of regressi no. Endo- MDP Dose Incidence ( .) Ex- Tempo- (Duration) Comple toxin analogue (nmol) - tent ral

Light Dar.

2 + Saline 5 60 0 37±4 60 (4±2) 20

-- ^" DP(J) 60 5 0 100 70±4 ^C 20 ( 3 ) 80

-f ir 6 5 0 100 67±2 ^C 40 < >11±8) 60

+ II 2 5 20 80 57±4° 60 ( 4±1) 40

+ L2-MDP 60 5 0 100 71±5 ^C 20 ( 4 ) 80

+ 6 5 0 100 67±1 ^C 40 ( 4±1) 60

+ 2 5 40 60 63±3 ^C 60 ( 12±4) 40

+ MDP-Lys (LIB) 60 5 0 100 75±3 ^C 40 ( 8±4) ^" 60

+ 6 5 0 100 74±1 ^C 40 ( 6±1) 60

+ π 2 5 0 100 66±5° 60 ( >13±5) 40

+ B30-MDP 60 5 0 100 64±4 ^C 80 ( 6±2*) 20

+ 6 5 40 60 49±3 ^C 80 9±4) 20

+ 2 5 60 20 48±5 80 . 7±4) 20

- MDP(J) 60 4 0 0 25 14 ) 25

- L2-MDP 60 4 0 0 0 0

- MDP-Lys (LB) 60 4 0 0 50 2±1) 0

- B30-MDP 60 4 0 0 0 25

- Saline 4 0 0 0 0

BALB/c mice with 9 day old Meth A sarcoma transplants were injected IV with 3 μg endotoxin alone or mixed with graded doses of an MDP analogue in a total volume of 0.5 ml saline or 0.1% Nikkol in saline (MDP-Lys(L1 and B30-MDP) . Solutions of all MDP analogues have been prepared the day before use and stored at -20°C.

b, c, d

See legend to Table G.

Table I. Potentiation of endotoxin-induced tumour necrosis and regressio by B30-MDP and MDP-Lys(Ll8) in various concentrations of Nikkol or saline.

. b

Exp. T r e a t m e n t ^a n Necrosis Incidence (%) of regression no.

Endo¬ MDP Dose Nikkol Incidence(%) Ex- Tempo- (Dura- Co toxin analogue ral tion) (nmol) (%) Light Dark

3 + Saline 5 40 0 48±7 40 ( 9±8) 0

+ B30-MDP 60 3 5 0 100 66±l ^c 20 ( 8 ) 80

+ 60 0.1 5 20 80 57±2 60 (>13±6) 40

+ ιι 6 3 4 75 25 56±5 100

+ 6 0.1 4 50 50 55±6 100

- Saline 3 4 0 0 25 (12 ) 0

- Saline 0 4 0 0 0 0

4 + Saline 10 40 40 49±3 50 ( 4±4) 20

^; + MDP-Lys (L18) 60 0.1 10 0 100 73±2 ^C 50 ( 9±4) 50

+ ." 60 0 10 10 90 70±5 ^C 60 (>10±4) 40

+ 6 0.1 10 0 100 66±5 ^C 30 ( 2±0) 70

+ 6 0 10 0 100 71±3 ^C 40 (>11±6) 60

- 60 0.1 9 11 0 40 11 (>19 ) 0

- Saline 0 9 0 0 0 0

BALB/c mice with 9 day old Meth A sarcoma transplants were injected IV with 3 μg endotoxin alone or mixed with graded doses B30-MDP or MDP- Lys (L18) in a total volume of 0.5 ml saline finally containing 3 or 0.1 Nikkol. MDP-Lys (L18) was also administered in saline after short sonica tion. . _ j

' ' See legend to Table G.

Table J. Effect of MDP(L-L) on the induction of tumour necrosis and regression by endotoxin.

Exp. T r e a t m e n t n N e c r o s i s Incidence (%) of re gression no.

Endo- MDP Dose Incidence(%) Ex- Tempo- (Dura- Comp toxin analogue (nmol) tent ral tion)

Light Dark

5 + Saline 5 80 0 44±1 40 (2+1) 40

+ MDP(F) 60 5 0 100 66±3 ^C 20 ( 4 ) 80

+ MDP(L-L) 60 5 80 0 35±7 80 (>9±4) 0

- MDP(F) 60 5 20 0 39 40 (5+3) 0

- MDP(L-L) 60 5 0 0 0 0

- Saline 5 20 0 35 0 0

6 + Saline 5 40 20 45±5 60 (2±0) 40

+ MDP(F) 60 5 0 100 62±3 ^C 20 ( 3 ) 80

+ ^■ MDP(L-L) 200 5 80 0 47±4 60 (2+0) 20

- Saline 5 0 0 0 0

BALB/c mice with.9 day old Meth A sarcoma transplants were injected IV with 3 μg endotoxin alone or mixed with an MDP analogue in a total volume of 0.5 ml saline b, c, d

See legend to Table G.

Table K. Effect of DDA and L121 on the induction of tumour necrosis and regression by endotoxin.

Exp. T r e a t m e n t n N e c r o s i s Incidence (%) of gression no.

Endo- Adjuvant Dose Incidence (%) Ex- Tempo- (Dura- Com toxin (nmol) tent ral tion)

Light Dark

7 + Saline 5 60 20 44+6 60 ( 1±0)

+ MDP(F) 60 5 20 80 71+3° 40 (>10±7)

+ DDA 160 5 100 0 44+3 40 ( 1±0)

- DDA 160 4 25 0 33 25 ( 1 ⁾

- Saline 4 0 0 0

+ Saline 10 60 10 42±5 50 (> 6±4)

+ MDP(F) 60 5 20 80 69+2° 40 (>11±8)

+ L121 25 10 80 20 49±4 80 (>10+3)

+ L121 75 5 80 20 55±3 ^C 80 ( 0±4)

- L121 25 9 0 0 0

- L121 75 5 0 0 100 ( 9±2)

( L121 25

+ ( 5 20 80 63±3 ^C 0 1 ( MDP(F) 60

( L121 25

— ( 5 60 0 38+3 0 ( MDP(F) 60

— Saline 10 0 0 10 ( 1 )

BALB/c mice with 9 day old Meth A sarcoma transplants were injected IV with 3 μg endotoxin alone or mixed with L121, DDA or MDP in 0.5 ml sal

' ' See legend to Table G.

Results

Effect of MDP and analogues on endotoxin-induced necrosis and regression.

IV administration of 3 μg endotoxin to tumour-bearing mice caused moderate necrosis and temporal and complete regre sion in 60% and 20% of the tumours respectively (Table G) . All MDP analogues induced only incidental necrosis and regres sion. Addition of either MDP analogue to endotoxin resulted i extensive dark stained necrosis of all tumours. Their potency to enhance the incidence of subsequent regression rather vari B30-MDP in 3% Nikkol was extremely potent, MDP and L2-MDP wer active to a lesser extent and MDP-Lys (L18) in 3% Nikkol had n effect at all, when compared to treatment with endotoxin alon To allow a possible distinction between the analoques, differ doses of the agents were compared for their potentiating ac¬ tivity (Table H) . With all agents a decrease of the dose resu ed in decreased potentiation of tumour damage. MDP, L2-MDP an MDP-Lys (L18) in 0.1% Nikkol appeared to be equally active but B30-MDP frozen in 0.1% Nikkol had only little activity. As th discrepancies in activity of MDP-Lys (L18) and B30-MDP in ex¬ periments 1 and 2 might be due to changes in Nikkol concentra tion and/or storage conditions, we studied the effect of thes analogues prepared freshly in different concentrations of Nik (Table I) . The effect of 60 nmol B30-MDP in 3% Nikkol was as high as observed in experiment 1 (Table G) while reduction of the Nikkol concentration to 0.1% resulted in the decreased activity as observed in experiment 2 (Table H) , A dose of 6 nmol B30-MDP in either 3 or 0.1% Nikkol potentiated necrosis moderately but caused maximal potentiation of regression. Fresh preparations of MDP-Lys (Ll8) in 0.1 Nikkol or saline ha about equal activity (Table I) ,

The effect of MDP (L-L) on endotoxin-induced tumour damage is shown in Table J. MDP (L-L) lacked any potentiating activity for necrosis and even tended to counteract inductio of regression by endotoxin.

Effect of DDA and 121 on endotoxin-induced necrosis and regression .

In further experiments we studied whether the synthet adjuvants DDA and L121 could replace MDP as to potentiation of endotoxin-induced tumour damage (Table K) . Both adjuvants caused little tumour damage of their own and hardly potentia endotoxin-induced tumour necrosis. The incidence of endotoxi induced tumour regression, however, was enhanced by addition of L121. Combined administration of L121 and MDP resulted in moderate necrosis and incidental regression. Endotoxin com¬ bined with the latter two agents caused extensive necrosis a complete cure in all mice treated. This treatment, however, was very toxic as judged by the profound lethargy and hypo¬ thermia of the animals.

Discussion

The MDP analogues L2-MDP and MDP-Lys (18) appeared to be as good potentiators of endotoxin-induced tumour necrosis as MDP. On molar base B30-MDP in 3% Nikkol had less activity in this respect (Tables G-I) . On the other hand the latter agent applied in a suitable concentration of Nikkol was the best potentiator of regression, followed by MDP and L2-MDP which had about equal activity, whereas MDP-Lys (L18) was the least potent. MDP (L-L) lacked any potentiating activity (Tabl J) . These observations may indicate that more or less distinc properties of MDP are involved in potentiation of endotoxin- induced necrosis and regression and are in keeping with the findings that induction of tumour necrosis and regression are separate events. Regarding the immune-independence and immune dependence of the respective phenomena of tumour damage, it is tempting to suggest that nonspecific properties of MDP and analogues are involved in potentiation of necrosis and that specific immunostimulating properties are involved in potenti tion of regression. The very prompt appearance of necrosis al most excludes a role of specific immunostimulating mechanisms

in potentiation of necrosis. This is supported by the observa tion that the non-MDP adjuvants DDA and L121 are almost devoi of potentiating activity. Stimulation of nonspecific bacteria resistance seems to be a common property of the most effectiv potentiators of tumour necrosis. The complete incapability of B30-MDP, however, to stimulate nonspecific resistance while its capacity to potentiate tumour necrosis is still consider¬ able, suggests that these parameters are not clearly related. A relation between the capacity to activate macro hages and to potentiate tumour necrosis also seems to be absent, as the macrophage activating properties of B30-MDP and DDA are in disproportion with their capacity to potentiate necrosis.

Concerning endotoxin-induced tumour regression the bes potentiator, B30-MDP, seems to be the most potent im uno- adjuvant of all MDP analogues tested, as in aqueous solution it stimulates induction of both antibody formation and delay¬ ed hypersensitivity. Whether the ability to stimulate' specifi cell-mediated immunity is related to the ability to potentiat endotoxin-induced tumour regression is questionable as: (1) DDA did not potentiate regression (Table K) although its pattern of immunoadjuvant action resembles that of B30-MDP an the incidence and extent of necrosis induced by DDA was not very different from that in the experiments with 6 nmol B30- MDP (Tables I and K) ; (2) MDP and various analogues in aqueous solution do not stimulate delayed hypersensitivity, but are powerful potentiators of tumour regression; (3) it is doubtful whether B30-MDP stimulates cell-mediated immunity when administered separate from the antigen (tumour) as induction of delayed hypersensitivity with MDP ^' in oil usual¬ ly required admixture with antigen. On the other hand it cannot be excluded that adjuvanticity for antibody formation is mechanistically involved in potentiation of tumour re¬ gression. MDP and analogues in aqueous solution can stimulate specific antibody formation even when given by other routes than the antigen. Furthermore, an other unrelated pure

humoral adjuvant, L121, induced a rather prolonged state of tumour growth inhibition upon combined administration with endotoxin, although this was not followed by a clearly en¬ hanced incidence of complete cures (Table K) . The latter may be due partly to the incapacity of L121 to potentiate necrosis, which may facilitate subsequent regression. However, a role of enhanced antibody formation in the potenti tion of endotoxin-induced tumour regression seems not obvious as regression was shown to require an intact T cell immune system and the presence of concomitant immunity, which is thought to be mediated by T cells alone or by T cells in con¬ junction with macrophages . Some role of specific anti-Meth A antibodies can, however, not be excluded in view of the follo ing data. Histologically confirmed delayed skin reactivity to Meth A could be transferred with serum from mice immunized with viable Meth A cells and this was suggested to be mediate by antibody. Moreover anti-Meth antibodies were shown to in¬ hibit the growth of Meth A in vitro as well as in vivo. For these reasons it seems useful to test whether administra¬ tion of MDP together with endotoxin to Meth A-bearing mice results in an increased amount of specific antibodies and if so whether this may play a role in the increased incidence of regression .

The efficacy of the lipophilic MDP analogues B30-MDP and MDP-Lys (L18) appeared to depend on addition of an optimal dose of the nonionic surfactant Nikkol. This might be related to the biological availability of these agents. In this respect study of the effect of incorporation of lipophilic MD analogues in liposomes deserves attention, as Fidler et al . (Proc .Natl .Acad. Sci . USA 78, 1680, 1981) have shown their use fulness in the nonspecific eradication of spontaneous metas- tases of Bl6 melanoma in C57BL/6 mice.

In conclusion we found that MDP and various analogues have the capacity to potentiate endotoxin-induced tumour ne¬ crosis and regression and that these effects cannot be attri¬ buted, as yet, to some other known biological effects of thes

agents .

Example IV

Older Meth A tumors are commonly insensitive to the induction of tumor regression by a high dose of endotoxin

(Berendt et al. , J.Exp.Med. 148, 1550 and 1560, 1978) . This is attributed to the induction of T suppressor lympho¬ cytes by larger i munogenic tumors (North, Adv.Immunol. 35, 89-155, 1984) . The failure to induce complete regressions in 15-day old tumors by the administration of 25 μg toxic endo¬ toxin was confirmed (Bloksma et al., Eur.J.Cancer Clin.Oncol. 20, 397-403, 1984? Table L , although necrosis induced is considerable. Data from Table L show that the combination of a low dose of toxic endotoxin with MDP not only induces con¬ siderable necrosis, but also incidental complete regressions of 15-day old Meth A tumors. A pretreatment of the tumor carrying animals on day 13 with a high dose of cyclophospha- mide (3 mg/mouse = 150 mg/kg) , followed by a combination of toxic or detoxified endotoxin and MDP on day 15, led to a high incidence of complete regressions.

Cyclophosphamide by itself caused retarded growth only. A lower dose of cyclophosphamide (0.9 mg/mouse) had no effect on tumor growth and did not potentiate the anti-tumor effect of toxic endotoxin with MDP. Still a lower dose of cyclo¬ phosphamide (0.3 mg/mouse) , which would rather selectively eliminate T suppressor lymphocytes, had no potentiating effect on the activity of toxic endotoxin with MDP either

(data not shown) .

TABLE

Effect of pretreatment with cyclophosphamide on the antitumor of MDP combined with toxic or detoxified endotoxin on 15-da Meth A tumors.

Treatment on Tumor necrosis Growth In< :ide

Exp. day day Incidence Extent ³ inhibition com no. 13 15 cu

1 Cy(3mg) Tox(3 _μ g)/MDP 5/5 72+2 >14 4/

Cy(3mg) Tox(3 _y g) 5/5 45±6 >13+1 0/ sal ne Tox(3 _u g)/MDP 5/5 75±2 7±0 2/ saline Tox(3 _y g) 5/5 43±2 l±l 0/ saline Tox(25 _v g) 5/5 73±4 4±3 0/

2 Cy(3mg) Tox(3 _μ g)/MDP 5/5 78±2 5/

Cy(0.9mg) Tox(3 _μ g)/MDP .5/5 75±3 > 9±2 0/ sal ne Tox(3 _μ g)/MDP 5/5 69±6 > 7±2 1/5

Cy(3mg) Detox(10 _μ g)MDP 5/5 63±10 12 4/5

Cy(3mg) saline 4/5 46±4 0±0 ^C 0/5

Cy(0.9mg) saline 5/5 43±7 O±O 0/5 saline saline 5/5 43±4 O±O 0/5

BALB/c mice with 13-day old Meth A tumors on the abdomen were injected with cyclophosphamide (Cy) or saline. On day 15 they were treated i.v. toxic endotoxin from E. coli 0111.B4 (Tox) or detoxified endotoxin fr typhimurium Re (Detox) combined with 30 _u g MDP in saline. ^a (Ratio of necrotic area and tumor area) x 100. (Mean ± SEM).

No increase of tumor size after treatment of mice which show no com cure (Mean number of cays z SEM). ^c Growth rate of these tu ors is markedly retarαec compared to saline-tr control s .

Ψ- l≠ §

-30-

Example V

Data from the following Tables M and N show that combinations of MDP with endotoxins are more effective than are endotoxins alone against two lymphoid tumors. In mice cured of their MOPC 315 tumor, metastases were observed a short time later, mainly in the inguinal and axillary lymph glands. These metastases were probably not induced by the treatment, as the surgical removal of 13-day old MOPC 315 tumors still fully localized in the skin at that time led to at least an equally high frequency of metastasis.

TABLΓ M

Effect of endotoxin/MDP against s.c. transplanted 8-day old SL2 tumors i syngeneic DBA/2 mice

Treatment i.v. Tumor necrosis Growth Incidence of on day 8 Incidence Extent ² inhibition complete cure

Tox/MDP 5/5 80±5 - 5/5

Tox 5/5 62±3 ^C 2±1 2/5

MDP 5/5 53±4 ^C O±O 0/5 saline 5/5 52+4 O±O 1/5

DBA/2 mice were injected s.c. with 1x10 viable SL2 cells (T cel lymphosarcoma) on the abdomen. On day 8 (tumor diameter + 7-8.5 mm) they wer injected i.v. with endotoxin from E. coli 0111:B4 (3 μg) and/or MOP (30 μg) i saline. ^a (Ratio of necrotic area and tumor area)x 100. (Mean ± SEM).

No increase of tumor size after treatment of mice which show no complet cure (Mean number of days ± SEM).

Intensity of necrosis is weak.

TABLE N

Effect of combinations against 12-day old MOPC 315 plasmacytomas in syng BALB/c mice

Exp. Treatment i.v. Necrosis Growth Incidence of no. on day 12 Incidence Extent ^b inhibition complete cur

1 Tox(10 _μ g)/MDP ^a Tox(25 _y g) ^a

Tox 10 _u g 5/5 81±3 5±3 l/5 ^d

Detox 10 _u g/MDP 5/5 91±3 3±0 3/5 ^d

Detox 10 _y g 3/5 52±9 O±O 0/5

MDP 5/5 52+3 0x0 0/5

2 Tox(3 _u g)/MDP 5/5 83±6 5+3 2/5 ^e

Tox(3 _u g) 5/5 56±5 0+0 0/5

Detox(10 _u g)/MDP 5/5 . 65+9 l±O 1/5

Detox 10 _μ g 5/5 51±3 O±O 0/5

MDP 4/5 45±3 O±O 0/5

BALB/c mice were injected s.c. with 1x10 viable M0PC315 cells on the abd Day 12 (tumor diameter ± 12.5 mm) they were treated i.v. with toxic endo from E. coli 0111.B4 (Tox) or detoxified endotoxin from S. typhimuriu (Detox) mixed with 30 _u g MDP in saline.

These treatments are lethal to all mice. (Ratio of necrotic area and tumor area)x 100. (Mean ± SEM).

No increase of tumor size after treatment of mice which show no defi cure (Mean number of days ± SEM).

These mice showed distant metastases after cure of the primary tumor.

One of these "ice showed distant rπetasTase-. after cure of the ?r- ^' tumor.