EP0732938A1

EP0732938A1 - Method of treating intestinal disorders

Info

Publication number: EP0732938A1
Application number: EP95902712A
Authority: EP
Inventors: Graham Douglas Fischer Jackson
Original assignee: Unisearch Ltd
Current assignee: Unisearch Ltd
Priority date: 1993-12-01
Filing date: 1994-12-01
Publication date: 1996-09-25
Also published as: JPH09505812A; WO1995015179A1; EP0732938A4; CA2177570A1

Abstract

The present invention provides a method of treating or preventing in a subject an intestinal disorder caused by an elevated level of TNF in the lumen of the intestine of the subject. The method involves administering to the subject an agent which reduces the action of the intralumenal TNF or reduces the production or accumulation of intralumenal TNF. The agent may be any of a number of known anti-TNF agents, however, it is preferred that the agent is an anti-TNF antibody. It is also preferred that the agent is administered directly into the intestine.

Description

METHOD OF TREATING INTESTINAL DISORDERS The present invention relates to a method of treating intestinal disorders. The method involves the use of procedures or agents that reduce the production or block the action of tumour necrosis factor α (TNF), produced in the liver and released into the bile of humans or animals. Since TNF released into the bile may in some circumstances be taken up from the gut and act systemically, the invention also relates to a method of treatment of disorders due to elevated circulating TNF, by blocking TNF release to or action in the bile.

The major symptoms of endotoxic shock are considered to be due to a range of pro inflammatory mediators with most attention focusing on tumour necrosis factor α (TNF), interleukin 1 (IL-1) and platelet activating factor (PAF) . How these agents might interact to produce the various effects is not known. Intestinal lesions are a common feature of endotoxaemia and both TNF and PAF have been implicated. Further, it is implicit in such studies that these agents operate solely from the circulation. The present inventors hypothesised that intestinal damage in endotoxic shock results from the action of lumenal agents. Since lipopolysaccharide (LPS) is removed in the liver, and given that Kupffer cells can produce the relevant cytokines, and that products synthesised in the liver are likely to appear in bile in at least trace amounts, the present inventors reasoned that LPS-induced hepatobiliary factors could directly interact with, and cause damage to the gut. The present inventors have now found that external biliary drainage of rats given a lethal dose of LPS prevents gastrointestinal lesions.

In addition, the present inventors have demonstrated that infusion of recombinant human TNF into the bile diverted from and then reintroduced to the gastrointestinal tract, results in similar symptoms to those observed in animals challenged with LPS. Furthermore, addition of antibodies directed against murine TNF (which also recognise rat TNF) to the bile of LPS-treated animals partially or fully blocked the gastrointestinal lesions seen in animals not treated with antibody, including haemorrhage and fluid accumulation, diarrhoea and gross intestinal damage. As well, the constitutional symptoms associated with LPS treatment and elevation of bile TNF levels (fever, hypotension, general malaise and tissue damage) were reduced. Accordingly, in a first aspect the present invention consists in a method of treating or preventing in a subject an intestinal disorder caused by an elevated level of TNF in the lumen of the intestine of the subject the method comprising administering to the subject an agent which reduces the action of the intralumenal TNF or reduces the production or accumulation of intralumenal TNF.

In a second aspect, the present invention consists in a method of treating or preventing in a subject constitutional symptoms resulting from increased uptake into the circulation of TNF produced by the liver and secreted into the bile, the method comprising administering to the subject an agent which reduces the production or action of biliary TNF. There are a large number of syndromes or conditions which include intestinal disorders amongst their symptoms. These syndromes or conditions include sepsis, diarrhoea (viral, bacterial), diarrhoea (induced by stress, alcohol, antibiotics, chemotherapy, or radiation), gay bowel syndrome, graft versus host disease (e.g., after bone marrow transplantation), colon cancer, HIV infection, AIDS and AIDS related infections (e.g., icrosporidium) , colic, ulcers initiated by non-steroidal anti-inflammatory drugs (NSAIDS), inflammatory bowel diseases (Crohns disease, ulcerative colitis, irritable bowel syndrome), graft versus host disease, gastrointestinal ulcers, gastrointestinal inflammation, bacterial translocation following surgery, multi-organ failure and food allergy. TNF has been implicated in many of these syndromes or conditions (9-18) . Also the importance of bile in the development of NSAIDS-induced intestinal damage has been previously demonstrated (19). Accordingly, in an embodiment of the present invention the intestinal disorder in the subject is as a result of one or more of these syndromes or conditions. In particular the subject is suffering from sepsis, diarrhoea (viral, bacterial), diarrhoea (induced by stress, alcohol, antibiotics, chemotherapy, or radiation), HIV infection, AIDS and AIDS related infections (e.g., Microsporidium), bacterial translocation following surgery, and bowel ischemia/reperfusion injury

A number of these conditions or syndromes, particularly those involving infection, cause an increase in the level of TNF in the bile which flows into the intestine. In a preferred embodiment of the present invention the elevated level of TNF in the lumen of the intestine is of bile origin.

The agent which reduces the action of the intralumenal TNF or reduces the production or accumulation of the intralumenal TNF may be any of a number of agents such as:- a) antibodies to TNF or to its receptors, e.g., a TNF- blocking antibody; b) the soluble forms of TNF receptors; c) peptides or other molecules that block TNF action at its receptors; d) peptides that reduce TNF production; e) agents that block TNF production, such as thalidomide, anti-inflammatory inhibitors of 5-lipoxygenase and 5-cyclooxygenase, e.g., Tepoxalin (Johnson & Johnson), Tenidap (Pfizer),

CGP47 969A (Ciba Geigy) , agents that reduce TNF production by elevating cAMP levels including pentoxyfy11ine; f) lipid molecules, such as polyunsaturated fatty acids that block TNF production. It is presently preferred that the agent is an antibody, either polyclonal or monoclonal. A preferred monoclonal antibody is MAb47. The hybridoma cell line which produces MAb47 was deposited with the European Collection of Animal Cell Cultures (ECACC), Vaccine Research Production Laboratory, Public Health Laboratory Service, Centre for Applied Microbiology and Research, Porton Down, Salisbury, Wiltshire SP4 OJG, United Kingdom on 14 December 1989 and was accorded Accession No 89121402. Other antibodies which may find use in the method of the present invention are those disclosed in International Patent Application Nos WO 90/01950, WO 89/08460 and WO 90/10707.

In a further preferred embodiment of the present invention, the agent is a peptide, preferably peptide 492 or peptide T. Further details regarding peptide 492 can be found in Australian Patent Application No 22731/92 "Peptide which abrogates TNF and/or LPS toxicity" . Further information regarding peptide T may be found in EP0249390, EP0249394 and WO88/09338. The disclosures of these references are included herein by reference.

An agent reducing the production of TNF could be delivered via a number of routes including: oral, parenteral, transdermal etc., or in combination with a second agent that would enhance accumulation of the agent in the liver. An agent reducing the action of biliary TNF could be delivered by a number of routes including:- a) direct delivery to a relevant region of the bile duct or gastrointestinal tract via a cannula or needle, or b) delivered orally and have its effect in the gut (i.e., formulated to be protected from degradation until released at the required site in the gut), or c) delivered parenterally and be excreted through the liver into the bile, or d) delivered parenterally and block transport of TNF to bile, or e) delivered via the rectum for lower bowel diseases, or f) any combination of these routes.

It is presently preferred that the agent is delivered directly to the gastrointestinal tract or bile duct.

In order that the nature of the present invention may be more clearly understood, a preferred form thereof will now be described with reference to the following examples and Figures in which:-

Figure 1 Volumes of clear intestinal fiuid in normal (a), LPS treated (b) , normal with bile duct cannulation (BDC) (c) LPS treated with BDC (d) and LPS treated with bile duct ligation (BDL) (e) rats. Volumes are shown as mean values ± S.E. (n=5)

METHOD: The small intestine of each rat was tied at each end and freed from the abdominal tissues and the contents expressed by gentle squeezing from the duodenum to ileum. The solid material, including the mucus, was removed by centrifugation (500 g, 15 min) and the volume of clear fluid measured.

Figure 2 Levels of biologically active TNF in the sera of Salmonella infected rats before, or 20hr after, treatment with 1ml of polyclonal rabbit anti-TNFα which was administered either intra-duodenally (I/Duo) or intravenously (I/V).

METHOD: Sera were titrated for TNF by a cytotoxicity assay using a TNF sensitive human rhabdomyosarcoma cell line (KYM-1)(20). EXAMPLE 1

When rodents are subjected to lethal doses of LPS, a series of well described events such as fever, hypotension, general malaise, and tissue damage precede death. The intravenous injection of 15mg/kg of

Escherlchla coll B04:lll LPS into adult Wistar rats results in 100% mortality 12-24hrs later. Examination at 4hrs post inoculation revealed macroscopic evidence of haemorrhage and fluid accumulation within the small bowel. Neither of these events were apparent in similarly treated rats which had had the common bile duct cannulated to prevent bile from entering the gut. No intestinal damage was apparent in normal animals or in bile duct cannulated animals (not treated with LPS). Less haemorrhage and no fluid accumulation was observed in bile duct-ligated, LPS-treated rats. Measurement of the volume of fluid present in the gut at this time revealed a significant difference between that present in LPS-treated rats and in those where bile was prevented from entering the gut (Fig. 1). To further implicate bile in these events, the bile ducts of LPS-treated rats were cannulated and the bile deviated to the upper jejunum. Here the duodenum appeared as normal, while haemorrhage and fluid accumulation occurred in the lower reaches of the intestine. Histological examination of tissues revealed that bile deprivation prevented the gross intestinal damage in LPS-treated rats. While neutrophil margination was quite apparent in intestinal blood vessels of both intact and BDC rats, only in the former group was there evidence of haemorrhagic events in the villi. Occasional villi in the BDC group had some oedema in the tips though clearly there was no major disturbance in the integrity of the tissues. It was observed, albeit subjectively, that the BDC group, even though they had undergone surgery were much more lively during the 4 hour period than their intact counterparts - this study was, however, not designed to determine the effect of bile deprivation on survival. It was also observed that stool of BDC animals was well formed.

Measurement of TNF (by bioassay) in the bile and sera of normal and LPS-treated animals revealed a substantial increase in biliary TNF levels in the treated group (Table 1) .

TABLE 1 TNFα Levels (ng/ml) in Serum and Bile of Rats injected IV with 15mg/kg E.coli LPS

TIME (hrs) SERUM BILE

0 ND 0-6

1.5 1-20 1000-3000

In this model of endotoxic shock, tissue damage also occurs in the stomach. This was observed in the above experiment at both macroscopic and microscopic levels.

However, no such damage was detected in the BDC animals. Thus it is proposed that reflux of bile containing raised levels of TNF can induce ulcer formation in the stomach.

EXAMPLE 2

Detection of high levels of tumour necrosis factor α(TNF) in the bile of rats given a lethal dose of bacterial lipopolysaccharide (LPS) and which causes severe intestinal damage, prompted experiments to directly test the hypotheses that: a) TNF is responsible for intestinal damage and b) that In situ neutralisation of TNF activity prevents the intestinal damage. To test hypothesis a) recombinant human TNFα in sterile saline was infused into the duodenum of a group of rats at a rate of lml/hr for 4 hr

0-30 1 μg/ml

30-60 5 μg/ml

60-120 10 μg/ml

120-180 5 μg/ml

180-240 1 μg/ml

A second group of animals was injected IV with LPS

(15mg/Kg as in Example 1), bile duct cannulated and infused with TNFα as above.

A further group of animals was infused with physiological saline (controls). After 4 hours the animals were killed and H&E stained sections of duodenum, jejunum and ileum were prepared.

Macroscopic and microscopic examination of the intestinal tract of both TNF-infused groups of animals showed a pattern of damage similar to that which occurred in rats injected with LPS and in which bile was allowed to flow normally. Saline infusion did not affect the gut. These results are consistent with hypothesis a).

To test hypothesis b) a group of rats which had been injected IV with LPS (15mg/Kg as in Example 1) was infused intra duodenum with Genzyme rabbit polyclonal anti-mouse TNFα (batch Nos B4377 and B4805).

A further group of LPS treated animals was infused with normal rabbit serum (NRS) . The rate of infusion of serum was lml/hour and at a dilutions (in saline) as follows:

0-30 1:10 dilution

30-60 1:5 dilution

60-120 1:2 dilution

120-180 1:5 dilution

180-240 1:10 dilution At four hours animals were killed and intestinal tissues sectioned and stained as above.

Tissues from the animals infused with anti-TNFα serum appeared as normal. No haemorrhage occurred at any site although occasional villi in the duodenum and ileal regions showed a small degree of oedema. No breaching of the epithelium was observed. The intestinal tracts of the rats infused with normal serum were damaged in the same fashion (necrosis, severe haemorrhage, villus damage or heavy oedema) as for non-infused animals injected with LPS. These results are consistent with hypothesis b) .

Further it was observed from histological examination of stomach tissue that the intraintestinal infusion of anti-TNF antibodies also protected the stomach from damage. EXAMPLE3 Liver macrophaσes as the source of biliary TNF

A further experiment showed that biliary TNF most likely arises from the Kupffer cells of the liver and also further implicates TNF as the damaging agent following LPS treatment.

Thus, rats treated with gadolinium chloride (which is known to eliminate Kupffer cells (21) 24 hr prior to treatment with the lethal dose of LPS are protected from intestinal damage. At the same time it was found that TNF levels in the bile of gadolinium chloride treated rats were not raised above the background level. This strongly supports the inventors contention that inhibition of TNF production in the liver is an alternative/further means of preserving the integrity of intestinal tissue.

EXAMPLE 4

The influence of bile TNF in the pathogenesis of an enteric pathogen Background: Current thinking and research on the invasiveness of an enteric pathogen is centred on those properties of the microorganism (virulence factors) which enhance the processes of adhesion to the epithelial surface, their proliferation and subsequent invasion into the local tissues . Little attention has been given to host factors (in particular inducible factors) which influence this process. The present inventors observations relating to the role of biliary tumour necrosis factor α in the intestinal effects of bacterial lipopolysaccharides (LPS, endotoxin) a constituent of Gram-negative bacterial pathogens, prompted experiments which tested the hypotheses that a) biliary factors induced by LPS alter the intestinal epithelium to permit early entry of organisms through this barrier and b) in the later stages of an infection, bile (and in particular the tumour necrosis factor α component) is responsible for the development of diarrhoea.

Model of infection: Rats if given an infective (non lethal) dose (10 -10 ) of live Salmonella enteriditis (Se) orally, show signs of disease from days 8-12 and are fully recovered by day 21. Rats given a high dose (of the order of 10 ) of organisms become sick within a week and if not treated would die at days 10-12. At days 5-6 the diarrhoea phase begins . The infection of rats with Se is regarded as one of the standard animal models of human typhoid fever.

Experiments:

1. Low dose infection of normal rats resulted in the target organs (liver and spleen) being invaded by day 3 at the earliest. However when rats were pretreated intravenously with low doses of LPS (l-300μg), which simulates the absorption of the Se LPS from the gut, invasion occurred earlier, and with doses above lOμg/rat within 24 hours. Importantly, bile duct cannulation ie. external biliary drainage of rats at the time of infection and LPS treatment (all doses) prevented appearance of Se in target organs within 24 hours. A further set of experiments determined that these doses of LPS transiently (for less than 16hr) increased gut permeability, offering the opportunity for pathogens to enter irrespective of whether they may eventually cause disease. Indeed, bacteria of the gut normal microbiota were detected along with the invasion of Se in these experiments. These very organisms are the prime candidates for the induction of bacterial sepsis in those circumstances where the gut is known to be compromised. These organisms also failed to gain entry (ie. bacterial translocation) if bile was diverted. Since other experiments have demonstrated the increase in biliary TNF after LPS treatment and that the gut retains its integrity if either bile is diverted from the intestine or if the TNF component is neutralised, it is believed that in these experiments gut permeability and the enhanced invasiveness is due to raised levels of biliary TNF.

2. External bile drainage of rats given the high dose (10 10) of Se when diarrhoea is obvious (day 6) resulted in the cessation of diarrhoea within 20 hr, whereas control animals continued to have diarrhoea. Assessment (eg. alertness etc) of the rats' condition after cannulation indicated an improved health status as compared to the non-cannulated group. Cannulation reduced the volume of intestinal fluid significantly and stool became formed. To evaluate the role of TNF in the diarrhoeal process, groups of rats were treated as follows:

(a) Injected IV with 1 ml of neutralising polyclonal rabbit anti-TNF antibody (Genzyme) .

(b) Infused in the upper duodenum with 1 ml of the above antiserum (diluted to 1/20) at a rate of lml/hr.

(c) Injected IV with 1 ml of normal rabbit serum (control) .

(d) Infused (as above) with 1 ml of diluted normal rabbit serum (control).

The experiment was terminated at 20hr.

Results:

Control groups (c) and (d) animals were indistinguishable in terms of health status and diarrhoea, i.e., signs of endotoxaemia (fur ruffling, loss of interest in food, inactivity) , with excessive diarrhoea.

Of the group (a) animals 50% died within the 20hr period. The survivors were visibly distressed, showing overt signs of endotoxaemia and frank diarrhoea (worse than Groups (c) and (d) .

Group (b) animals were assessed to be in better health than Groups (c) and (d) . Diarrhoea had ceased in all animals .

Evidence of cessation of diarrhoea in group (b) is provided by a statistically significant reduction in intestinal fluid volume as compared to the volumes present in the intestine of animals in the other groups (Table 2) . Table 2 Volume of intestinal fluid in S. enteritidls infected rats after infusion of anti-TNF antibody

Infusion Volume (ml) p value

(mean±ISD) (^as compared to control infected) ^nil 5.88+1.63

intraintestinal 2.30±0.62 0.006 anti-TNF intravenous 4.2811.19 0.16 anti-TNF intraintestinal 5.95±1.07 0.94

NRS intravenous NRS 5.07±1.28 0.47

The difference between the intraintestinal anti-TNF and intravenous anti-TNF groups was statistically significant (p=0.026).

It was also noted that stool was more formed in animals of group b) . Also it is noted that group b) animals had received 20ml intraintestinally yet volume was significantly less than volume in other groups

Group (b) animals had significantly lower numbers of Se in the mesenteric lymph nodes than group (a) animals. Numbers of organisms in Peyers patches of group (b) animals were also lower but not statistically different from numbers in group (a) (Table 3). Table 3

Numbers of S. enteritidls in rat tissue 20 hr after treatment with anti-TNF antibodies

(Mean No. of organisms (logιo)/g™ tissue ± ISD)

Tissue Group a Group b p value

Mesenteric 6.89±0.05 6.32+0.29 -⁰⁰⁸ lymph nodes

Peyers 7.22±0.23 7.03±0.21 0.271 patches

Histological examination of intestinal tissues revealed that a substantial improvement in the integrity of the epithelium had occurred in group b animals. In addition serum TNF levels were reduced to undetectable levels in these animals (Fig 2).

Conclusions:

The results of these experiments are consistent with the hypotheses as stated above. The particular merit of these experiments is that inhibition of TNF activity (by diversion of bile from the gut or In si tu neutralisation) in an established diarrhoeal condition dramatically terminates the episode. Earlier experiments (LPS toxic model) demonstrated that intraintestinal anti-TNF therapy prevented the initiation of intestinal damage.

In contrast to intra-duodenal administration of anti-TNF antibody, systemic delivery of the same dose of antibody failed to protect the animals and, instead, the constitutional symptoms of endotoxaemia progressed more rapidly in this group of animals . These results of the effects of systemic delivery of anti-TNF antibodies are consistent with the findings of other investigators (2-6), and may help to explain why recent clinical trials with systemically administered anti-TNF antibodies and other TNF inhibitors have produced disappointing results (7) . EXAMPLE 5 Study to assess the levels of TNF in the faeces of humans with various intestinal disorders

Faecal samples were prepared for assay by published methods. The assay system - Enzyme Linked Immunoabsorbent Assay - was established according to a published protocol (8) .

Patients with the following conditions were found to have statistically (using the Kruskal-Wallis One Way Analysis of Variance) significant higher levels of faecal TNF than levels present in faecal material from normal individuals.

Inflammatory Bowel Diseases (Confirming published data (9) ) . Human Immunodeficiency Virus/AIDS patients with a positive diagnosis of Microsporidia . Bone Marrow transplantation patients with diagnosed graft versus host disease.

It is concluded that these results are consistent with the view that intestinal disturbances correlate with the presence of raised levels of TNF.

As can be seen from the results of Examples 1-5, the present inventors have demonstrated a strong correlation between gastrointestinal lesions and elevated levels of TNF in the bile. Furthermore, they have effectively demonstrated that addition to the bile of an agent able to block the action of TNF also blocks the associated gastrointestinal lesions.

The present inventors have also observed a reduction in general constitutional symptoms (fur ruffling. disinterest in food, inactivity, isolation from other animals) associated with elevated bile TNF levels, either by diversion of the bile from the gut following cannulation of the bile duct, or in LPS-treated animals given antibody to TNF into the gastrointestinal tract. Many toxaemias in man, including that associated with the septic shock syndrome, are associated with excessive production of TNF in a number of tissues and organs including the liver. A number of publications report a causal associated between TNF production and various pathological symptoms. We suggest that in man, as in animals, the release of TNF from the liver, via the bile duct into the gastrointestinal tract (G.I.) or the direct release of TNF into the intestinal lumen from other sources, is the direct cause of a number of endotoxaemias including septic shock and herein describe a method for preventing the effects of the production in the liver of a toxic amount of TNF on the G.I. tract and its consequences in establishing serious disease conditions.

It is our contention that stimuli such as the presence in the systemic circulation of bacterial cell components such as the bacterial cell wall lipopolysaccharide fraction (LPS) induces production of massive amounts of TNF in the liver. This TNF is secreted via the bile duct into the duodenum where it causes an inflammatory response in the region of the G.I. tract distal to its point of entry. In turn, this causes damage to the lining of the gut wall including breeching its integrity and allows contents of the G.I. tract to diffuse into the systemic circulation. These contents include other bacteria and bacterial cell components which then initiate a further round of induction of TNF in the liver, which in turn is released into the gut and keeps the cycle going, with serious consequences to the health of the individual. We have devised a number of methods which terminate this disease cycle and cure the symptoms of the disease. These procedures which may be seen as alternatives, or could be used in concert to achieve a more effective result, are:

1. The prevention of the secretion of TNF, produced in the liver, via the bile duct into the G.I. tract. This can be accomplished simply by insertion of a cannula into the bile duct by any standard procedure to enable the content of the bile duct including its TNF to be drained to the exterior. This drainage would continue until the symptoms of the disease resolved.

2. Neutralisation of TNF in the G.I. tract. The bile duct discharges its contents into the upper part of the G.I. tract in the duodenum adjacent to and distal to the pyloric sphincter of the stomach. The contents of the duodenum are approximately neutral in pH, and antibodies delivered directly into the duodenum are reasonably stable. Consequently, one effective way of neutralising TNF in the duodenum is to inject an antibody able to bind TNF directly, for example, via insertion of a stomach tube into the lumen of the gut passing through the stomach and the pyloric sphincter directly into the duodenum, or by passing a needle through the duodenal wall from the exterior guided by a keyhole surgical device. An appropriate antibody would be dripped continuously into the duodenum via one or other of these tubes to maintain an excess over the amount of TNF present so as to inactivate it, until such time as the release of TNF from the liver dropped to normal limits . In the experiments, which are described above, heterologous polyclonal antibodies produced in rabbits were injected into the experimental animal, which in this case was the rat, which had been injected with LPS to elicit a severe toxaemia. In treating human toxaemias we would prefer to use a polyclonal antibody which could be produced by hyperimmunising a variety of animals with human TNF by standard procedures, or manufactured by mixing together a number of monoclonal human antibodies . The animal sources of the polyclonal anti-TNF antibodies could be horses, cows, sheep, camels, goats or rabbits. Monoclonal antibodies could also be effective. Whilst human monoclonal antibodies able to block the biological activity of human TNF would be less likely to produce immunological side-effects, in practice, monoclonal antibodies from other animals delivered into the G.I. tract could be as effective. Apart from delivering antibodies or other therapeutic agents able to inactivate TNF directly into the duodenum or other parts of the G.I. tract distal to the duodenum and where the gut wall damage occurs, it is obvious that indirect inactivation methods could also be effective. Also, anti-TNF antibody, or other TNF inhibitors could be delivered orally in a suitable formulation, to be released at the required site. In some cases, inflammation may be caused by localised production of TNF by TNF secreting cells in the gut (10, 11). A number of simple formulations and coatings are in use to protect labile pharmaceuticals from degradation in the stomach whilst allowing the active component to be available lower down the G.I. tract. The use of such coatings and/or formulations could have benefit to enable the oral therapeutic use of antibodies, peptides, etc., for treatment of the less traumatic toxaemias where the patient is conscious and able to swallow tablets and where the urgency for a rapid resolution of the disease is not so pressing. It is also obvious that in suggesting that these endotoxaemias could be treated by delivering therapeutic agents in the upper region of the G.I. tract, via the stomach, or a tube inserted through the body wall, that similar therapeutics could also be delivered to those parts of the G.I. tract most affected by exposure to TNF, via a tube inserted through the rectum into the lower bowel to prevent TNF- related damage to regions of the bowel adjacent to or distal to the site of discharge of the antibody.

In addition to antibodies able to neutralise TNF in the G.I. tract, other therapeutic agents able to modulate the function of TNF and to diminish its toxic effects, could be used to treat such toxaemias by delivering them into the gastrointestinal tract in the same way as we have described for antibodies . Examples of such compounds are a number of peptides which are able to suppress the binding of TNF to its receptors . Peptides derived from TNF or from TNF receptors and their analogues have been shown to have activity in blocking TNF function and would be useful for this therapeutic role. Others, such as Peptide T (EP0249390, EP0249394 and WO88/09338) are able to block certain functions of

TNF and could also be used when delivered in the fashion indicated.

In human therapy it may be necessary to deliver the therapeutic antibody or peptide on a continuing basis directly into the duodenum or other regions of the GI tract. Since patients with a serious toxaemia are generally comatose, this is most conveniently accomplished by the use of a stomach tube passed into the duodenum. Antibody would be continuously dripped into the duodenum at a rate which allowed an excess of that necessary to completely inactivate TNF passing into the duodenum from the bile duct. This effect can be achieved in a practical sense by periodic sampling, via the stomach tube of the fluid in the duodenum for assay for the level of free TNF and conversely of free anti-TNF antibody.

Whilst we believe that either cannulation of the bile duct, or G.I. tract therapy by compounds inactivating the effects of TNF, would have benefit when used alone, the combined use of both procedures would ensure a more certain therapeutic outcome. The precise cimount administered may be determined as in the procedure described above.

While the use of the present invention has been described with particular reference to the treatment of humans it will be readily appreciated that the method of the present invention is also equally applicable to the treatment of animals other than humans .

It will be appreciated by persons skilled in the art that numerous variations and/or modifications may be made to the invention as shown in the specific embodiments without departing from the spirit or scope of the invention as broadly described. The present embodiments are, therefore, to be considered in all respects as illustrative and not restrictive.

REFERENCES

1. Tran-Thi, T.-A., Weinhold, L., Weinstock, C, Hoffmann, R. , Schulze-Specking, Northoff, H., Decker, K. (1993). Production of tumour necrosis factor-a, interleukin-1 and interleukin-6 in the perfused rat liver. Eur. Cytokine Netw. , 4 , 363. 2. Eskandari M.K., Bolgos, G., Miller, C, Nguyen, D.T., DeForge L.E., Remick, D.G. (1992). Anti-tumor necrosis factor antibody therapy fails to prevent lethality. J. Immunol. 148, 2724. 3. Ecthenacher, B., Falk, W. , Mannel, D.N., Krammer,

P.H. (1990). Requirement of endogenous tumour necrosis factor/cachectin for recovery from experimental peritonitis. J. Immunol. 145, 3762. 4. Bagby, G.J., Plessala, K.J., Wilson, L.A., Thompson, J.J., Nelson, S. (1991). Divergent efficacy of antibody to tumor necrosis factor/cachectin for recovery from experimental periotonitis. J. Infect. Dis . 163, 83. 5. Zanetti, G. , Heumann, D., Gerain, J., Kohler, J., Abbet, P., et al, (1992) Cytokine production after intravenous or peritoneal Gram-negative bacterial challenge in mice. Comparative protective efficacy of antibodies to tumor necrosis factor α and to lipopolysaccharide. J. Immunol., 148, 1890.

6. Nakane A., Minagawa, T., Kato, K. (1988). Endogenous tumor necrosis factor (cachectin) is essential to host resistance against Listeria monocytogenes infection. Infect. Immun. 56, 2563.

7. Morrison, D.C., Dinarello, C.A., Munford, R.S., Natanson, C, Danner, R. , Pollack, M. , Spitzer, J.J., Ulevitch, R.J., Vogel, S.N., McSweegan, E. (1994). Current status of bacterial endotoxins . ASM News, 60, 479.

8. Danis, V.A. , Franic, G.M. , Rathjen, D.A. , Brooks, P.M. (1991). Effects of granulocyte-macrophage colony- stimulating factor (GM-CSF), IL-2, interferon-gamma (IFN- γ) , tumour necrosis factor-alpha(TNF-α) and IL-6 on the production of immunoreactive IL-1 and TNF-α by human monocytes. Clin. exp. Immunol, 85, 143. 9. Braegger, C.P., Nicholls, S.W., Murch, S.H.,

Stephens, S., MacDonald, T.T. (1992). TNFα in stool as a marker of intestinal inflammation. Lancet, 339 , 89. 10. Tan, X., Hsueh, W. , Gonzalez-Crussi, F. (1993). Cellular localisation of tumor necrosis factor (TNF)- alpha transcripts in normal bowel and in necrotising enterocolitis. TNF gene expression by Paneth cells, intestinal eosinophils, and macrophages . Am. J. Pathol. 142, 1858.

11. MacDonald, T.T. , Murch, S.H., Nicholls, S.W., Breese, E.J. (1993). Diarrhoeal disease: current concepts and future challenges . Intestinal cytokines in inflammatory bowel disease and invasive diarrhoea. Trans of the Royal Soc. of Trop. Med. & Hyg. 87, Suppl. 3, 23.

12. Braegger, C.P., MacDonald, T.T. (1994). Immune mechanisms in chronic inflammatory bowel disease. Ann. Allergy 72, 135.

13. MacDonald, T.T., Hutchings, P., Choy, M.-Y., Murch, S., Cooke, A., (1990). Tumor necrosis factor-alpha and interferon-gamma production measured at the single cell level in normal and inflammed human intestine. Clin. exp. Immunol, 81 , 301.

14. Olson, A.D., Ayass, M. , Chensue, S. (1993). Tumor necrosis factor and IL-lβ expression in pediatric patients with inflammatory bowel disease. J. Pediatr Gastroenterol Nutr. 16, 241.

15. Caputi, A., Squadrito, F. (1992). Role of TNF-a and therapeutic perspectives in bowel and myocardial ischemia/reperfusion injury. Pharmacol Res. 26, 150.

16. Arnold, J.W. , Niesel, D.W. , Annable, C.R., Hess, C.B., Asuncion, M. , Cho, Y.J., Peterson, J.W. , Limpel, G.R. (1993). Tumor necrosis factor-α mediates the early pathology in Salmonella infection of the gastrointestinal tract. Microb. Pathog. 14 , 217.

17. Piquet, P.-F., Grau, G.E., Allet, B., Vassalli, P. (1987). Tumor necrosis factor/cachectin is an effector of skin and gut lesions of the acute phase of graft-vs.- host disease. J.Exp. Med. 166, 1280.

18. Bemelmans, M.H.A., Gouma, D.J., Greve, J.W., Buurman, W.A. (1993). Effect of antitumour necrosis factor treatment on circulating tumour necrosis factor levels and mortality after surgery in jaundiced mice. Br. J. Surg. 80, 1055. 19. Fang, W.-F., Broughton, A, Jacobson, E.D., (1977). Indomathecin-induced intestinal inflammation. Dig. Dis . 22, 749. 20. Meager, A., (1991). A cytotoxicity assay for tumour necrosis factor using a human rhabdomyosarcoma cell line. J. Immunol. Methods 144 , 141. 21. Hardonk, M.J., Dijkhuis, F.W.J., Hulstaert, C.E., Koudstaal, J., (1992). Heterogeneity of rat liver and spleen macrophages in gadolinium chloride-induced elimination and repopulation. Journal of Leukocyte Biology 52, 296.

Claims

CLAIMS : -

1. A method of treating or preventing in a subject an intestinal disorder caused by an elevated level of TNF in the lumen of the intestine of the subject the method comprising administering to the subject an agent which reduces the action of the intralumenal TNF or reduces the production or accumulation of intralumenal TNF.

2. A method as claimed in claim 1 in which suject is suffering from a syndrome or condition selected from the group consisting of sepsis, diarrhoe, gay bowel syndrome, graft versus host disease, colon cancer, HIV infection, AIDS and AIDS related infections, colic, ulcers initiated by non-steroidal anti-inflammatory drugs, inflammatory bowel diseases, graft versus host disease, gastrointestinal ulcers, gastrointestinal inflammation, bacterial translocation following surgery, multi-organ failure and food allergy.

3. A method as claimed in claim 2 in which the suject is suffering from a syndrome or condition selected from the group consisting of sepsis, diarrhoea, HIV infection, AIDS and AIDS related infections and bacterial translocation following surgery.

4. A method as claimed in any one of claims 1 to 3 in which the TNF in the lumen of the intestine of the individual is of bile origin.

5. A method as claimed in any one of claims 1 to 4 in which the agent is selected from the group consisting of:- a) antibodies to TNF or to its receptors, e.g., a TNF- blocking antibody; b) the soluble forms of TNF receptors; c) peptides that block TNF action at its receptors; and d) combinations thereof.

6. A method as claimed in any one of claims 1 to 5 in which the agent is a polyclonal or monoclonal antibody.

7. A method as claimed in any one of claims 1 to 6 in which the agent is monoclonal antibody MAb47. 7. A method as claimed in any one of claims 1 to 6 in which the agent is monoclonal antibody MAb47.

8. A method as claimed in any one of claims 1 to 5 in which the agent is a peptide, preferably peptide 492 or peptide T.

9. A method as claimed in any one of claims 1 to 7 in which the agent is delivered by a route selected from the group consisting of:- a) direct delivery to a relevant region of the bile duct or gastrointestinal tract via a cannula or needle; b) delivered orally'and have its effect in the gut; c) delivered parenterally and be excreted through the liver into the bile; d) delivered parenterally and block transport of TNF to bile; and e) a combination thereof.

10. A method as claimed in any one of claims 1 to 9 in which the agent is delivered directly to the gastrointestinal tract or bile duct.

11. A method as claimed in claim 3 in which the agent reduces the production of biliary TNF.

12. A method as claimed in claim 11 in which the agent is selected from the group consisting of:- a) peptides that reduce TNF production; b) agents that block TNF production, such as thalidomide, anti-inflammatory inhibitors of 5-lipoxygenase and 5-cyclooxygenase, agents that reduce TNF production by elevating cAMP levels including pentoxyfylline; and c) lipid molecules, such as polyunsaturated fatty acids that block TNF production.

13. A method of treating or preventing in a subject constitutional symptoms resulting from increased uptake into the circulation of TNF produced by the liver and secreted into the bile, the method comprising administering to the subject an agent which prevents the production or action of biliary TNF.

14. A method as claimed in claim 13 in which the agent is selected from the group consisting of:- a) antibodies to TNF or to its receptors, e.g., a TNF- blocking antibody; b) the soluble forms of TNF receptors; c) peptides that block TNF action at its receptors; and d) combinations thereof.

15. A method as claimed in claim 13 or 14 in which the agent is a polyclonal or monoclonal antibody.

16. A method as claimed in any one of claims 13 to 15 in which the agent is monoclonal antibody MAb47.

17. A method as claimed in any one of claims 13 to 15 in which the agent is a peptide, preferably peptide 492 or peptide T.

18. A method as claimed in any one of claims 13 to 17 in which the agent is delivered by a route selected from the group consisting of:- a) direct delivery to a relevant region of the bile duct or gastrointestinal tract via a cannula or needle; b) delivered orally and have its effect in the gut; c) delivered parenterally and be excreted through the liver into the bile; d) delivered parenterally and block transport of TNF to bile; and e) a combination thereof.

19. A method as claimed in any one of claims 13 to 18 in which the agent is delivered directly to the gastrointestinal tract or bile duct.

20. A method as claimed in claim 13 in which the agent is selected from the group consisting of:- a) peptides that reduce TNF production; b) agents that block TNF production, such as thalidomide, anti-inflammatory inhibitors of 5-lipoxygenase and 5-cyclooxygenase, agents that reduce TNF production by elevating cAMP levels including pentoxyfylline; and c) lipid molecules, such as polyunsaturated fatty acids that block TNF production.

21. A method as claimed in any one of claims 1 to 20 in which the subject is a human.