WO2005111605A2 - Biomarkers for the prediction and treatment of drug-induced diarrhoea - Google Patents

Abstract

The invention provides biomarkers for the prediction of diarrhoea in a subject to whom a microtubule-stabilizing agent has been administered, based upon the gene expression of certain biomarker genes by the subject.

Description

BIOMARKERS FOR THE PREDICTION AND TREATMENT OF DRUG-INDUCED DIARRHOEA

FIELD OF THE INVENTION

[0001] This invention relates generally to the analytical testing of tissue samples in vitro, and more particularly to the analysis of gene expression profiles or haematology profiles as biomarkers for predicting drug-induced diarrhoea.

BACKGROUND OF THE INVENTION

[0002] Epothilone B (EPO906) is a natural product produced by the myxobacterium Soragium celleulosum (Bollag DM et al, Cancer Res. 55(11): 2325-33 (1995)) that is currently being studied as single- agent therapy against many forms of solid tumours. The epothilone B mechanism of action is similar to that of the taxane family of cytotoxics. Epothilone B acts by promoting microtubule polymerization that leads to a mitotic block in the cell cycle at the G2-M phase transition, ultimately leading to apoptotic cell death. Rothermel J et al., Semin. Oncol. 30(3 Suppl 6):51-5 (June 2003). An advantage of epothilone B over the taxane class of antiproliferation drugs is that epothilone B is equally cytotoxic to drug-sensitive and multidrug-resistant cells overexpressing P-glycoprotein, even against taxol-resistant cancer cells.

[0003] With no myelosuppression having been observed to date, epothilone B-induced diarrhoea is the dose-limiting toxicity. Rothermel J et al, Semin. Oncol. 30(3 Suppl 6):51-5 (June 2003). Drug-induced diarrhoea is not unique to epothilone B. Diarrhoea has been reported for a variety of anticancer drugs targeted to inhibit the cell cycle, such as CPT-11 and paclitaxel. Trifan OC et al, Cancer Res. 62 (20):5778-84 (2002); Mavroudis D et al, Oncology 62 (3):216-22 (2002).

[0004] There is a need in the art to increase the safety and efficacy of epothilone B anticancer therapy in individual patients by predicting whether the patients will experience drug- induced diarrhoea and by targeting appropriate therapies to the individual patients.

SUMMARY OF THE INVENTION

[0005] The invention provides methods for predicting diarrhoea in a subject to whom a microtubule stabilizing agent has been administered. The invention also provides methods for treating diarrhoea in a subject to whom a microtubule stabilizing agent has been administered. In one embodiment, the methods involve the steps of obtaining the gene expression profile of the subject, wherein the gene expression profile comprises the gene expression pattern of one or more genes predictive of the occurrence of diarrhoea in a subject following administration of a microtubule stabilizing agent, and determining whether the subject is at risk for developing diarrhoea from the administration of the microtubule stabilizing agent. The microtubule stabilizing agent can be an epothilone or an epothilone analogue. [0006] The subject can be a mammal. While the EXAMPLES below describe the administration of epothilone B to a rat, the results can reasonably be extended to other mammals, including humans, because of the similarity of mammalian immunological and digestive physiology.

[0007] The invention further provides a model for targeting an appropriate anti-diarrhoeal intervention. The results provided herein show that administration of epothilone B produces a pronounced, early increase in the expression of these inflammatory genes, which remains high over time. Initially, destabilized membrane phospholipids release pro-inflammatory molecules and cytokines, like platelet activating factor (PAF) and tumour necrosis factor (TNF). Consequently, endothelial activation and initiation of the inflammatory cascade include the generation of phospholipase A₂ (PLA₂) and hence, synthesis of prostaglandins occurs. Among the consistently highly expressed genes at the early time points that were determined to be involved in the manifestation of diarrhoea in the high-dose animals were proinflammatory molecules (phospholipase A₂ (PLA₂) and the complement component Clq), cytokines and their receptors (interleukin 1 alpha (ILlα) and Tumour Necrosis Factor Receptor 1 (TNF- Rl)), neuromediators (alpha2D, alpha2A adrenergic receptors and tachykinin 2), and some transporters like aquaporine 8. The occurrence of diarrhoea and inflammation is concomitant to the increase expression of tumour necrosis factor 1 (TNFR1) mRNA and phospholipase A2 (PLA₂) mRNAs. The triggering of TNFR1 and the prostaglandin pathway, in particular PLA₂ (phospholipaseA2) appear to be involved in inducing diarrhoea. There is also concomitant increase expression of the P-glycoprotein/multidrug resistance 1 (mdrl) gene. Later, the inflammatory effect of the compound is superposed to its pharmacological action on tubulins. [0008] The genomic results provided herein, combined with the clinical and pathological findings of multifocal necrosis (erosions) at 96 hours, support the hypothesis that the loss of the water/electrolyte barrier function was a major factor for diarrhoea. Hence, the dose and time dependent increase of the inflammatory response incremented the degree of cell damage and resulted in diarrhoea. Thus, loss of the cryptal cell architecture by cell death and necrosis and hence, the water/electrolyte balance function of the gut is responsible for the diarrhoea. These results suggest that a therapeutic intervention could be most promising if initiated as soon as possible, before induction of acute phase proteins by EPO906 administration, to counteract the effect of translational processing into the protein version of some of these mRNAs such as TNFα, cathepsin C and PLA₂ and further downstream effects. [0009] The invention also provides clinical assays, kits and reagents for predicting diarrhoea in a subject to whom a microtubule stabilizing agent has been administered. In one embodiment, the kits contain reagents for determining the gene expression of certain genes, where the expression profile of the genes is a biomarker for the risk of the subject for experiencing diarrhoea.

BRIEF DESCRIPTION OF THE DRAWINGS

[0010] FIG. 1 is a diagram showing the genes in the prostaglandin pathway triggered by

EPO906.

[0011] FIG. 2 is a diagram showing the timing of the events of the gene changes induced by EPO906 in the rat caecum.

[0012] FIG. 3 is a bar graph showing the mean sPLA₂ enzymatic activity in serum samples of control and 0.4 mg/kg EPO906 treated animals.

[0013] FIG. 4 is a bar graph showing the mean sPLA enzymatic activity in serum samples of control and 1.75 mg/kg EPO906 treated animals.

[0014] FIG. 5 is a graph showing a typical standard curve for sPLA₂ enzymatic activity.

[0015] FIG. 6 is a diagram showing an inflammatory mechanism triggering diarrhoea in the caecum of the Lewis rat treated with EPO906. Destabilized membrane phospholipids release proinflammatory molecules and cytokines like PAF (platelet activating factor) and

TNF. Consequently, endothehal activation and initiation of the inflammatory cascade include the generation of PLA₂ and synthesis of prostaglandins. At 24 hours, the inflammatory action of the compound is superposed to its pharmacological action on tubulins, and elevation of the mdrl gene (multidrug resistance protein 1). DETAILED DESCRIPTION OF THE INVENTION

[0016] The invention advantageously provides a way to determine, during the course of treatment, whether a patient will experience diarrhoea during drug treatment. [0017] As used herein, a gene expression profile is predictive of the occurrence of diarrhoea when the increased or decreased gene expression is an increase or decrease (e.g., at least a 1.5-fold difference) over the baseline gene expression following administration of a microtubule stabilizing agent. Alternatively, a gene expression profile is also predictive of the occurrence of diarrhoea when the increased or decreased gene expression correlates significantly with subjects who develop drug induced diarrhoea and/or the lack of increased or decreased gene expression correlates significantly with subjects who do not develop drug induced diarrhoea. Accordingly, in one embodiment, the invention used pharmacogenomic diagnostics as part of a therapeutic intervention. Pharmacogenomics is the correlation of a patient's genetic information to drug response, leading to a "personalized" prescription of the most appropriate medicine, known as "personalized medicine" to those of skill in the art. See, Pharmacogenomics- A Personalized Approach to Medicine (LeadDiscovery Ltd., 2002). One of the approaches used in personalized medicine, rather than administering a therapeutic intervention to each member of a patient population, is instead identifying a subpopulation of patients in the patient population, the member of the subpopulation being those patients who are likely to benefit from the therapeutic intervention, thus increasing the therapeutic effectiveness and efficiency.

[0018] In one embodiment, the standard biomarker gene expression profile is the gene expression profile or average of gene expression profiles of a vertebrate to whom epothilone B has been administered, this profile or profile being the standard to which the results from the subject following administration is compared. Such an approach, which contains aspects of therapeutics and diagnostics, is termed "theranostic" by many of those of skill in the art. [0019] In a particular embodiment, the vertebrate is a mammal, such a rat. In a more particular embodiment, the mammal is a primate, such as a monkey or a human. As used herein, the administration of an agent or drug to a subject or patient includes self- administration and the administration by another. [0020] As used herein, a gene expression pattern is "higher than normal" when the gene expression (e.g., in a sample from a treated subject) shows a 1.5-fold difference (i.e., higher) in the level of expression compared to the baseline samples. A gene expression pattern is "lower than normal" when the gene expression (e.g., in a sample from a treated subject) shows a 1.5-fold difference (i.e., lower) in the level of expression compared to the baseline samples.

[0021] As used herein, a change in gene expression is "early" when it occurs by about four hours or less, for example, by about 2 hours.

[0022] Furthermore, the invention provides biomarkers predictive of the occurrence of diarrhoea in a subject to whom a microtubule stabilizing agent. See, e.g., TABLE 18. [0023] Thus, these genes and markers are useful as biomarkers for the prediction of diarrhoea by monitoring gene expression, e.g., in tissue from the digestive system, after drug treatment. See, e.g., TABLE 2, TABLE 12 and TABLE 20. In a preferred embodiment, the sample tissue is from the caecum.

[0024] These results can reasonably be extrapolated to the prediction of diarrhoea in patients following the administration of any diarrhoea-inducing microtubule stabilizing agent or derivative thereof, based upon the structural similarity or the modes of action in the gut of microtubule stabilizing agent to epothilone. See, Su et al, Angew. Chem. bit. Ed. Engl. 36(19): 2093-2096 (1997) and Chou et al, Proc. Natl Acad. Sci. USA 95: 9642-9647 (August 1998). The microtubule stabilizing agent may be an epothilone or an analogue. U.S. Pat. Appln. 20030114450. Among the epothilones and epothilone derivatives are those described in U.S. Pat. Nos. 5,969,145, 6,583,290 and 6,605,726; U.S. Pat. Applns. 20020028839 and 20030114450; PCT patent publications WO 99/54330, WO 99/54319, WO 99/54318, WO 99/43653, WO 99/43320, WO 99/42602, WO 99/40047, WO 99/27890, WO 99/07692, WO 99/02514, WO 99/01124, WO 98/25929, WO 98/22461, WO 98/08849, and WO 97/19086; and German Pat. No. DE 41 38 042. In a preferred embodiment of the invention, the microtubule stabilizing agent is epothilone B or an analogue thereof, such as BMS-247550. . [0025] Moreover, the results can be extrapolated to the prediction of diarrhoea in patients who are being treated for diseases other than solid tumours. The method of the invention is applicable to vertebrate subjects, particularly to mammalian subjects, more particularly to human subjects. [0026] Techniques for the detection of gene expression of the genes described by this invention include, but are not limited to northern blots, RT-PCT, real time PCR, primer extension, RNase protection, RNA expression profiling and related techniques. Techniques for the detection of gene expression by detection of the protein products encoded by the genes described by this invention include, but are not limited to, antibodies recognizing the protein products, western blots, immunofluorescence, immunoprecipitation, ELISAs and related techniques. These techniques are well known to those of skill in the art. Sambrook J et al, Molecular Cloning: A Laboratory Manual, Third Edition (Cold Spring Harbor Press, Cold Spring Harbor, 2000). In one embodiment, the technique for detecting gene expression includes the use of a gene chip. The construction and use of gene chips are well known in the art. See, U.S. Pat Nos. 5,202,231; 5,445,934; 5,525,464; 5,695,940; 5,744,305; 5,795,716 and 5,800,992. See also, Johnston, M. Curr. Biol 8:R171-174 (1998); Iyer VR et al, Science 283:83-87 (1999) and Elias P, "New human genome 'chip' is a revolution in the offing" Los Angeles Daily News (October 3, 2003).

[0027] The synthesis and use of epothilones and epothilone derivatives are described in U.S. Pat. Nos. 5,969,145, 6,583,290 and 6,605,726; PCT patent publications WO 99/54330, WO 99/54319, WO 99/54318, WO 99/43653, WO 99/43320, WO 99/42602, WO 99/40047, WO 99/27890, WO 99/07692, WO 99/02514, WO 99/01124, WO 98/25929, WO 98/22461, WO 98/08849, and WO 97/19086; German Pat. No. DE 41 38 042; and scientific references cited therein.

[0028] As used herein, the administration of an agent or drug to a subject or patient includes self-administration and the administration by another.

[0029] The diagnosis of diarrhoea and other side effects of epothilone administration can be readily accomplished by those of skill in the medical arts. Rothermel J et al, Semin. Oncol. 30(3 Suppl 6):51-5 (June 2003). Diarrhoea may be treated with anti-diarrhoeal agents such as opioids (e.g. codeine, diphenoxylate, difenoxin, and loeramide), bismuth subsalicylate, and octreotide. Nausea and vomiting may be treated with anti-emetic agents such as dexamethasone, metoclopramide, diphenyhydramine, lorazepam, ondansetron, prochlorperazine, thiethylperazine, and dronabinol.

[0030] The maximum tolerated dose (MTD) for a compound is determined using methods and materials known in the medical and pharmacological arts, for example through dose-escalation tests. One or more patients is first treated with a low dose of the compound, typically 10% of the dose anticipated to be therapeutic based on results of in vitro cell culture assays. The patients are observed for a period of time to determine the occurrence of toxicity. Toxicity is typically evidenced as the observation of one or more of the following symptoms: vomiting, diarrhoea, peripheral neuropathy, ataxia, neutropaenia, or elevation of liver enzymes. If no toxicity is observed, the dose is increased 2-fold, and the patients are again observed for evidence of toxicity. This cycle is repeated until a dose producing evidence of toxicity is reached. The dose immediately preceding the onset of unacceptable toxicity is taken as the MTD. A determination of the MTD for epothilone B is provided above. [0031] The kits of the invention may contain a written product on or in the kit container. The written product describes how to use the reagents contained in the kit to determine whether a patient will experience diarrhoea during drug treatment. In several embodiments, the use of the reagents can be according to the methods of the invention. In one embodiment, the reagent is a gene chip for determining the gene expression of relevant genes.

[0032] The following EXAMPLES are presented in order to more fully illustrate the preferred embodiments of the invention. These EXAMPLES should in no way be construed as limiting the scope of the invention, as defined by the appended claims.

EXAMPLE 1

MICROARRAY mRNA EXPRESSION ANALYSIS IN RAT CAECUM

[0033] Summary. The purpose of this EXAMPLE was to better understand the pathogenesis of the diarrhoea after EPO906 administration, and evaluate strategies for the prevention or reduction of dose-limiting diarrhoea in patients. Caecum tissue from Lewis rats treated with a single intravenous dose of the novel microtubule-targeting agent epothilone B (EPO906) at 0.4 mg/kg and 1.75 mg/kg, were analyzed for changes in the mRNA expression levels at sampling times between 4 and 168 hours after treatment. [0034] Both dose and time dependent changes in the mRNA expression levels were observed reflecting mainly the pharmacological action of the compound on tubulin polymerization, stabilization of microtubules, cell cycle arrest, and induction of apoptosis. In addition, various genes indicative of an important inflammatory response and genes belonging to the acute phase protein family, coagulation cascade, tissue remodelling, angiogenesis, and lipid metabolism were differentially regulated. In particular, the changes in the expression of a number of genes involved in the prostaglandin pathway suggest that these could be important in the mechanism of diarrhoea. At the high dose, there was a pronounced, early increase in the expression of these inflammatory genes, which remained high over time, contrary to the low dose.

[0035] These results, combined with the clinical and pathological findings of multifocal necrosis (erosions) at 96 hours, support the hypothesis that the loss of the water/electrolyte barrier function was a major factor for diarrhoea. Hence, the dose and time dependent increase of the inflammatory response incremented the degree of cell damage and resulted in diarrhoea. Among the consistently highly expressed genes at the early time points and determined to be involved in the manifestation of diarrhoea in the high-dose animals were proinflammatory molecules (phospholipase A₂ (PLA₂) and the complement component Clq), cytokines and their receptors (Interleukin 1 alpha (IL1 alpha) and Tumour Necrosis Factor Receptor 1 (TNF-Rl)), neuromediators (α2D, α2A adrenergic receptors and tachykinin 2), and some transporters like aquaporine 8.

[0036] These results suggest that a therapeutic intervention could be most promising if initiated as soon as possible, before induction of acute phase proteins by EPO906 administration, to counteract the effect of translational processing into the protein version of some of these mRNAs such as TNFα, cathepsin C and PLA₂ and further downstream effects. [0037] Analysis design. Rats injected with EPO906 intravenously with a single dose of 1.75 mg/kg developed diarrhoea at day 4 to 5 after treatment while the low dose group of 0.4 mg/kg did not.

[0038] EPO906 was administered as a single intravenous (slow bolus) injection at 0.4 or 1.75 mg/kg using a volume-dosage of 1.14 and 5.0 mL/kg, to groups of 5 male Lewis rats, which were sacrificed after a recovery period of 4, 24 and 48 hours or 4 and 7 days. The animal species and strain used were LEW/Crl Ico rats from Charles River (F) Iffa-Credo SA, L'Arbresle, France. 55 male animals were used, aged 12 weeks (at dosing) with a body weight range of 321.0 to 375.7 g (at dosing). Five control animals similarly received a single intravenous administration of the reference item (placebo to EPO906 diluted with physiological saline) and were sacrificed 7 days after the administration. After a 14 day pretest period, a single intravenous dose was given, followed by a 4, 24 and 48 hours or 4 and 7 days recovery period. TABLE 1 Animal Allocation and Test Item Dosage- Group 1 2 3 2 3 2 3 2 3 2 3 Item Refer, EPO906 EPO906 EPO906 EPO906 EPO906 item Necropsy after 7 days 4 hours 24 hours 48 hours 4 days 7 days administration Sex Male Male Male Male Male Male Dose (mg kg) 0 0.4 1.75 0.4 1.75 0.4 1.75 0.4 1.75 0.4 1.75 Dosage 5.0 1.14 5.0 1.14 5.0 1.14 5.0 1.14 5.0 1.14 5.0 volume (mL/kg) Number of 5 5 5 5 5 5 5 5 5 5 5 animals

[0039] In vivo examinations and results. Data collections were performed individually. Clinical signs were assessed using a defined scoring system. Efforts were made to characterize onset and duration of signs observed. Special emphasis was placed on the observation of diarrhoea. Therefore, the animals were kept individually for approximately 3 hours per day from day 2. During this time, diarrhoea was recorded individually with the following grading: 1 = soft stool (still formed); 2 = soft stool (not formed); and 3 = liquid stool. For the observation of general clinical signs during group caging, diarrhoea was recorded as grade 1 (soft stool) or grade 2 (liquid stool).

[0040] All animals survived the scheduled test period. Clinical signs were limited to the animals at 1.75 mg/kg and consisted of piloerection from day 4, chromorhinorrhoea and chromodacryorrhoea from day 4 or 5 and diarrhoea from day 5. Diarrhoea on day 5 included soft (unformed) as well as liquid stool. On days 6 and 7, stool was generally liquid, with the exception of one animal (no. 53), which also showed soft (unformed) faeces. Animals at 1.75 mg/kg severely lost body weight (-26.1% of initial) within one week after the administration. Animals at 0.4 mg/kg generally showed a transient minimal to slight body weight loss with a trend to recovery at the end of the 1-week observation period.

[0041] Clinical pathology results. Blood samples were taken from the tongue vein during inhalation anaesthesia with Isoflurane (Forene^®; Abbott Laboratories S.A., Cham, Switzerland). Blood samples for haematology were collected into EDTA anticoagulant while those for clinical biochemistry were collected using no anticoagulant. [0042] Haematology changes at 1.75 mg/kg consisted of slight increases in red blood cell parameters (red blood cell count, haematocrit and haemoglobin) on days 5 and 8, probably due to fluid loss secondary to diarrhoea. Reticulocyte counts were slightly to markedly decreased from day 2 to 8, which appeared to be most pronounced on day 5. Slightly to moderately decreased platelet counts were noted on days 3 and 5, and markedly decreased platelet counts on day 8. Decreases in total white blood cells, neutrophils, lymphocytes, monocytes and/or eosinophils were generally evident from 4 hours after the administration to day 8. These reductions were most severe on day 5. In view of the percentage of white blood cells, neutrophils and monocytes showed a stronger reduction than lymphocytes on day 5. Increased numbers and percentage of large unstained cells were noted on day 8. Morphological changes in red blood cells on day 8 included the appearance of acanthocytes. Morphological changes in white blood cells on day 8 consisted of neutrophils or monocytes with basophilic plasm, hypersegmented neutrophils and atypical plasmatic lymphocytes. [0043] At 0.4 mg/kg, there were only minimal to slight transient effects including a decrease in reticulocytes on day 5.

[0044] Tissue sampling. At the following time points after the administration: 4, 24 and 48 hours, 4 and 7 days (see TABLE 1, above) animals were killed by exposure to carbon dioxide followed by exsanguination. [0045] The organs/tissues listed in TABLE 2 were taken from all animals. TABLE 2 Tissue list Sampling Histopathologv OCT frozen Genomics Cytokine release processing tissue measurements caecum m colon Ξ duodenum Ξ ileum Ξ jejunum m liver Ξ rectum Ξ

[0046] For histopathological processing, tissues were fixed in phosphate-buffered 4% formalin (w/v). Tissues were embedded in Paraplast^®, sectioned at (nominally) 4 microns and stained with haematoxylin & eosin.

[0047] Samples for the gene expression profiling were quick-frozen in liquid nitrogen immediately after excision, stored on dry ice and subsequently in a deep-freezer at approximately -80°C until further use.

[0048] Pathology results. All histologically processed tissues were subjected to microscopic examination. The pathological findings were multifocal necrosis (erosions)- and inflammation at 4 days. At 7 days, the mucosa was regenerating which was evident by an immature epithelium. Consequently, the comparison of the gene changes before 4 days might give insight into the mechanism of the diarrhoea at the high dose.

[0049] At necropsy, 2/5 rats had a thick caecum at 1.75 mg/kg 4 days after administration of the test item. Treatment-related microscopic findings were noted in the small and large intestine, and in the liver at 0.4 and 1.75 mg/kg, and after all recovery periods (from 4 hours to 7 days after the admimstration). Major microscopic findings for the caecum are summarized in TABLE 3. TABLE 3 Mai or microscopic findin igs in the caecum Dose 0.4 mg/kg 1.75 mg/kg Time after administration 4 h 24 h 48 h 4 d I 4 h 24 h 48 h 4 d 7_d Large intestine: Erosion (caecum) + + Increased mitotic figures + + (+) (+) + +(+) + (+) (+) + Apoptosis/s. cell necrosis + + (+) (+) (+) +(+) + (+) (+) Distorted architecture (rectum) + Diffuse infiamm. (mainly c.) + ++(+) ++(+) Focal peritonitis (caecum) + ++ Infiamm. oedema, submuc. (c.) + +(+) ++ Mucosal regeneration + + +(+) c: caecum s. cell necr.: single : cell necrosis (+):minimal +:minimal slight +(+):sl: ight/moderate ++:moderate ++(+):moderate/marked -H-+:marked

[0050] Immunohistochemistry results. The summary of the results of i munohistostaining of TNFα, IFN gamma and COX-2 in the caecum are presented in

TABLE 4. TABLE 4 Immunofluorescence investigations of cytokines and COX-2 in the caecum Dose 0.4 mg/kε 1.75 mg/kε Time after aclministration 4 h 24 h 48 h 4 d 7_d 4 h 24 h 48 h 4 d 7 d Caecum: TNF-α (mucosa) (+) + + (+) + + +(+) IFN-γ (mucosa) + (+) + (+) COX-2 + + + +) +: minimal/slight ++: moderate +++: marked

[0051] Increased TNFα concentrations were found in the caecum of single animals at 1.75 mg/kg on days 5 and 8. In general, there was a high variation of individual values. [0052] There was a dose-dependent increase in the mdrl gene expression of the caecum with highest mdrl mRNA levels on day 5. At 0.4 mg/kg, however, this effect was only minimal to slight, which attained statistical significance only on days 5 and 8. [0053] RNA extraction and purification. Total RNA was obtained by acid guanidinium thiocyanate-phenol-chloroform extraction (Trizol®, Invitrogen Life Technologies) from each frozen tissue section and the total RNA was then purified on an affinity resin (Rneasy®, Qiagen) according to the manufacturer's instructions and quantified. Total RNA was quantified by the absorbance at λ = 260 ran (A ₆on_m), and the purity was estimated by the ratio A _60nm/A₂so_nm- Integrity of the RNA molecules was confirmed by non-denaturing agarose gel electrophoresis. RNA was stored at approximately -80°C until analysis. One part of each individual RNA sample was kept for the analysis of critical genes by means of real-time PCR. [0054] GeneChip® assays. All GeneChip® assays were conducted according to the manufacturer's protocol (Affymetrix, Santa Clara, Calif, USA). The labelled RNAs were hybridized on Affymetrix rat RG U34A expression probe arrays. The data were stored for further export and loading into the GeneSpring® software 5.0 (Silicon Genetics, Calif, USA) used for analysis.

[0055] Double stranded cDNA was synthesized with a starting amount of approximately five μg full-length total RNA extracted as described above, using the Superscript Choice System (Invitrogen Life Technologies) in the presence of a T7-(dT) 24 DNA oligonucleotide primer. Following synthesis, the cDNA was purified by phenol/chloroform/isoamyl alcohol extraction followed by ethanol precipitation. The purified cDNA was then transcribed in vitro using the BioArray^® High Yield RNA Transcript Labeling Kit (ENZO) in the presence of biotinylated ribonucleotides form biotin labelled cRNA. The labelled cRNA was then purified on an affinity resin (Rneasy®, Qiagen), quantified and fragmented. An amount of approximately 10 μg labelled cRNA was hybridized for approximately 16 hours at 45°C to an expression probe array. The array was then washed and stained twice with streptavidin- phycoerythrin (Molecular Probes) using the GeneChip Fluidics Workstation 400 (Affymetrix). The array was then scanned twice using a confocal laser scanner (GeneArray® Scanner, Agilent) resulting in one scanned image. This resulting ".dat-file" was processed using the MAS 5 statistical algorithm (Affymetrix) into a ".eel-file". The ".eel file" was captured and loaded into the Affymetrix GeneChip Laboratory Information Management System (LIMS). The L S database is connected to a UNIX Sun Solaris server through a network filing system that allows for the average intensities for all probes cells (CEL file) to be downloaded into an Oracle database. Raw data were converted to expression levels using a "target intensity" of 150. The numerical values displayed were processed signal intensities of the probe-pairs comprised in a probe-set for a given transcript sequence. The data were checked for quality, exported from the database and loaded in the GeneSpring software 5.0 (Silicon Genetics, Calif, USA) for analysis.

[0056] Data analysis. Various filtering and clustering tools in these programs were used to explore the datasets and identify transcript level changes that inform on altered cell and tissue functions and that can be used to establish working hypotheses on the modes of action of the compound.

[0057] RNA samples were studied by using the human Affymetrix rat RG U34A GeneChip®. On such chip platform, probe-sets for individual genes contain 20 oligonucleotide pairs, each composed of a "perfect match" 25-mer and a "mismatch" 25-mer differing from the "perfect" match oligonucleotide at a single base. After probe labelling, hybridization, and laser scanning, the expression level is estimated by averaging the differences in signal intensity measured by oligonucleotide pairs of a given probe (AvgDiff value). The image acquisition and numerical translation software used for this EXAMPLE was the Affymetrix Microarray Suite version 5 (MAS5). The numerical values as shown in the TABLES below are a result of these estimations and were transferred for the analysis into the Silicon Genetics GeneSpring 5.0 software toolkit.

[0058] To identify genes that are impacted by treatment, the dataset is initially filtered to exclude in a first wave of analysis genes whose values are systematically in the lower expression ranges where the experimental noise is high (at least an AvgDiff value of 20 in a number of assays corresponding to the smallest number of replicas of any test point). In a second round of selection a threshold t-test p-value (0.05) identifies genes with different values between treated and non-treated for each group based on a two component error model (Global Error Model) and, where possible, with a stepdown correction for multi-hypothesis testing (Benjamini and Hochberg false discovery rate). Venn diagrams are used to identify the gene changes that are in common between the different groups.

[0059] The decision to consider a specific gene relevant was based on a conjunction of numerical changes identified by exploratory filtering and statistical algorithms as described above and the relationship to other modulated genes that point to a common biological theme. [0060] Increase and decrease in expression reported here are referring to the RNA expression levels unless specifically stated.

[0061] Gene expression profiling results. Pooled data from individual chips (5 rats in a group) are presented in the following TABLES. The significance of a gene expression change was a combination of fold change and results of statistical analysis. This was related to other modulated genes that point to a common biological theme. Extensive gene changes were induced by the treatment with EPO906 largely consistent with the known pharmacological and toxicological mechanism of actions.

[0062] Effect ofEPO906 on tubulins and cell death. At both doses, the gene changes reflected the pharmacological action of the compound on the cytoskeleton, i.e., the effect on the tubulins (TABLE 5) and the induction of cell death through caspase dependent and independent mechanisms (TABLE 6).

TABLE i Pharmacological action of EPO906 on tubulins Control EPO906. fold chang e* 7 davs 4 hours 24 hours 48 hours 4 davs 7 davs

AFFI ΓD Title AV SD 0.4 1.75 0.4 1.75 0.4 1.75 0.4 1.75 0.4 1.75 mg/kg mg/kg mg/kε mε/kε mε/kg mε kε mg/kg mε/kε mg/kε mε/k rc_AA800948_at Similar to tubulin alpha 256 46 1.3 1.6 1.7 2.3 1.2 1.6 1.0 1.0 1.5 1.4 V01227_s_at alpha-tubulin 613 61 1.4 1.6 1.4 3.2 1.3 3.1 1.2 3.0 0.7 1.3 rc_AA892333_at alpha-tubulin 2322 175 1.3 1.3 1.5 1.9 1.3 1.6 1.1 1.5 1.2 1.7 AB011679_at tubulin, beta 5 111 2 1.2 1.5 0.9 1.2 1.0 1.7 1.3 1.6 1.2 2.0 X03369 s at Rat mRNA for beta- 119 16 1.6 1.4 1.6 M 1.2 3 1.6 Λ5 1.1 5 tubulin T betal5 re AA799591 at Similar to tubulin T 380 27 1.2 1.3 1.5 L6 1.2 1.3 1.2 1.1 1.1 1.2 betal5

AB015946 s at tubulin, gamma 1 70 10 1.1 1.1 1.0 1.0 1.1 1.1 1.0 1.4 1.6 1.7 *Significant gene changes are bolded and underlined. AV: average SD: standard deviation

TABLE 6 Pharmacological action of EPO906 on cell death Control EPO906, fold change* y- 7 davs 4 hours 24 hours 48 hours 4 davs 7 davs

AFFI ΓD Title AV SD 04 1.75 04 1.75 04 1.75 04 1.75 04 1. mg/kg mg/kg mg kg mε/kε mg/kg mg/kg mε/kε mg/kε mε/kg m

AF025670 at caspase 6 112 13 1.0 0.9 0.9 0.7 0.7 0.8 1.0 1.1 1.0 1.

AF025670_g at caspase 6 219 23 0.8 0.8 0.9 0.8 0.7 0.6 0.7 1.0 1.0 1.3

U14647 at caspase 1 425 50 0.9 0.9 1.3 0.9 1.2 1.1 0.8 05 0.9 0.

U14647_g at caspase 1 177 34 0.9 0.9 1.4 0.9 1.2 1.1 0.8 05 0.9 0.

U49930 at caspase 3 78 19 1.1 1.1 1.4 OJ 0.9 08 0.8 ft5 0.9 0.

U49930 g at caspase 3 195 56 1.4 1.3 1.6 0.8 0.7 08 0.9 04 0.8 0.

U84410 s at caspase 3 117 19 0.9 1.0 1.2 0.5 0.6 0.5 Qά 03 0.5 0.

U77933 at caspase 2 316 45 1.0 0.9 0.9 0.8 0.9 08 0.8 0.7 0.9 0.

U75689_s_at deoxyribonuclease I- 147 33 1.3 1.2 1.3 1.5 1.6 M. 1.7 1.8 1.0 1. like 3

D25233UTR#l_g_ Retinoblastoma 1 126 17 0.8 07 1.1 06 0.9 05 0.7 04 0.9 0 at (including osteosarcoma)

M24604 at PCNA 1004 189 0.8 0.9 0.8 0.6 0.9 06 1.1 1.0 1.0 1.

D16308 at cyclin D2 322 74 1.1 0.9 1.6 13 2.6 hi M 5.4 2.3 2.

D16309_at CyclinD3 312 9 08 07 0.9 0 0.8 0.6 08 0.8 0.7 0.

D16309_g_at Cyclin D3 362 16 0.9 0.8 0.9 0.8 0.9 08 0.9 1.0 0.9 0.

D90404_at cathepsin C 325 53 1.4 1.6 1.1 1.7 1.1 2 1.8 4.4 1.0 3.

D90404_g at cathepsin C 131 7 0.9 1.3 1.0 1.6 1.1 3 0 1.4 6.9 1.0 4.

L13039 s at calpactin I heavy chain 1383 75 1.1 1.1 1.2 2.3 1.5 2.4 2.0 5.1 1.9 5.

U75689_s_at deoxyribonuclease I- 147 33 1.3 1.2 1.3 1.5 1.6 1.7 1.8 1.0 1. like 3

U24174_at cyclin-dependent kinase 90 15 1.1 1.2 1.2 1.5 1.0 1.6 1.1 27 1.0 1. inhibitor 1A

U66470_at cell growth regulatory 25 6 0.9 0.9 1.0 2.1 1.2 2 1.6 1.2 1.0 1. with EF-hand domain

L13039 s at calpactin I heavy chain 1383 75 1.1 1.1 1.2 2.3 1.5 2.4 2.0 5.1 LI 5

AF027954_at Bcl-2-related ovarian 381 33 0.9 08 1.2 0.7 0.9 OJ. QiZ 03 07 Q killer protein

TABLE 6 Pharmacological action of EPO906 on cell death Control EPO906. fold change* 7 davs 4 hours 24 hours 48 hours 4 davs 7 davs

AFFI ΓD Title AV SD 0.4 1.75 0.4 1.75 0.4 1.75 04 1.75 1.7 mg kg m /kg mg kg mg kg mg kg mg/kg mg kg mg kg mg/kε mg

AF027954 g_at Bcl-2-related ovarian 539 28 1.0 0.9 1.2 0.9 1.0 0.9 0.8 05 0.8 0.5 killer protein rc_AI008815_s_at cytochrome c, somatic 620 68 0.9 1.1 1.0 0.9 1.2 SA . 1.2 0.8

♦Significant gene changes are bolded and underlined.

[0063] Inflammatory effect ofEPO906. In addition, there was a remarkable effect of EPO906 on triggering genes involved at all levels of the inflammatory cascade, including complement, platelet activation, and macrophage activation, release of mediators, vascular events, tissue remodelling and repair (TABLE 7).

[0064] At the high dose, there was an early increase in the number of these genes and their expression levels remained high over time, contrary to the low dose.

TABLE 7 Selected genes in the inflammatory cascade induced by EPO906 Control EPO906, fold change* 7 davs 4 hours 24 hours 48 hours 4 davs 7 davs

AFFI ID Title AV SD 04 1.75 04 1.75. 0 1.75 04 1.75 04 1 mg/kg mg/kg mg/kg mg/kg mg/kg mg/kg mg/kg mg/kg mg kg m

VASCULAR EVENTS

U07619 at Coagulation factor HI 22 5 1.2 1.3 1.1 2.2 1.3 £0 1.8 M 1.0

M81642 at coagulation factor II receptor 409 116 1.0 1.4 09 1.5 1.0 1.7 3.1 0.8 1

M23697 at Plasminogen activator, tissue 46 19 1.4 1.6 0.8 1.7 1.0 3.2 1.7 16.0 0.6 1

M24067_at serine (or cysteine) proteinase inhibitor, 22 6 1.1 1.8 08 2J. 1.0 2.0 1.7 70 1.0 member 1

X63434 at Urinary plasminogen activator, urokinase 65 17 1.1 1.3 1.0 1.3 1.1 2.0 1.7 1 0.9

X71898_at Plasminogen activator, urokinase 110 20 1.1 1.4 1.3 1.5 1.2 2.0 1.4 IL 1.0 receptor

MACROPHAGE PRIMING

X05834 at Fibronectin 1 364 177 1.2 1.3 0.8 1.6 0.9 2.2 1.5 2.7 0.7 1

X52477 at Complement component 3 2 1 1.6 2.1 0.8 1.9 0.9 3.0 2.1 10.2 1.1 8

AB003042 at complement component 5, Rl 33 11 1.0 1.4 1.0 1.6 1.2 M. 1.3 14 1.0

U42719 at Complement component 4 281 132 1.2 1.5 0.9 2.5 0.8 3.6 1.9 5.4 0.7

M63122 at TNFRsuperfamily, member la 203 19 1.1 1.4 1.0 1.6 1.1 2.0 1.3 2.6 1.0 1

M98820 at Interleukin 1 beta 56 15 0.9 1.0 0.9 0.9 1.2 1.6 1.1 1.2 1.0

M98820 g at Interleukin 1 beta 36 6 1.1 0.9 0.9 0.9 1.2 1.6 1.1 1.2 0.8

D00403 at Interleukin 1 alpha 31 7 1.5 2.2 1.3 1.3 1.3 3.9 2.0 5.4 1.5

M34253 at Interferon regulatory factor 1 238 38 08 08 1.0 1.0 0.9 1.1 0.9 06 0.7

M34253 g at Interferon regulatory factor 1 561 73 0.7 0.8 1.1 1.1 1.1 1.3 09 0.7 0.8

AF053312 s at small inducible cytokine subfamily A20 568 260 09 0.7 1.1 0.5 0.9 1.9 1.2 0.2 0.5

Y08358 at small inducible cytokine subfamily Al 1 256 35 1.1 1.3 0.9 1.3 1.4 M 1.6 2.1 08 re All 80013 at Fc receptor, IgG, alpha chain transporter 1115 32 0.9 0.9 07 08 08 07 08 1.0 0.8

REPAIR re AA891828 g at procollagen, type I, alpha 2 103 8 0.9 0.7 08 0.9 09 0.9 0.9 07 08

X70369 s at collagen, type III, alpha 1 807 237 1.5 I A 0.9 1.4 1.1 M 1.8 33 0.9

X94551 at la inin, gamma 1 95 33 1.5 1.5 09 1.2 0.9 2.1 1.8 3.9 0.9

TABLE 7 Selected genes in the inflammatory cascade induced by EPO906 Control EPO906, fold change* 7 davs 4 hours 24 hours 48 hours 4 days 7 davs AFFI 1D Title AV SD 04 75 04 75 04 L75 04 175 QΛ L mε/kε mg/kg mg/kg mg/kε mg/kg mε/kε mε/kε mε/kε mg/kg m re AI104225 at Laimnin chain beta 2 91 34 1.3 1.3 0.8 1.7 1.0 2.4 1.7 2.7 0.7 1 U65656_at matrix metalloproteinase 2 153 32 1.2 1.4 1.0 2.0 1.1 3.1 1.7 8.7 0.9 AV: average SD: standard deviation *Significant gene changes are bolded and underlined.

>

[0065] Effect ofEPO906 on the prostaglandin pathway. Interestingly, the genes in the prostaglandin pathway leading to the arachidonic acid metabolites that are mediators of inflammation were well represented (TABLE 8).

[0066] Also at 1.75 mg/kg, as soon as 24 h, the mRNA for the P-glycoprotein/multidrug resistance 1 was significantly changed and remained high at all time points.

TABLE 8 Genes in the prostaglandin pathway affected bv EPO906 Control EPO906. fold change* 7 davs 4 hours 24 hours 48 hours 4 davs 7 davs AFFI H) Title AV SD 04 1.75 0.4 1.75 04 1.75 θ 1.75 04 1 mg/kg mg/kg mε kg mg/kε mg/kg mg/kg mg/kε mg/kε mg/kg m J03960_at arachidonate 5- 35 7 0.9 1.0 1.0 1.3 1.0 2.3 1.1 2.2 0.2 lipoxygenase M99567 at phospholipase C, beta 3 222 21 1.0 1.1 0.7 ^■ 0.6 0.5 0.4 0.6 03 0.7 M99567_g at phospholipase C, beta 3 126 2 0.9 1.1 0.7 0.6 0.6 0.4 0.5 03 0.7 X51529 at phospholipase A2, UA 1175 96 1.0 1.2 1.2 2.4 2.2 4.7 3.0 4.7 2.2 S67722_s_at prostaglandin-endoperoxide 16 2 1.4 1.4 1.3 1.4 1.2 1.8 1.4 £0 0.6 synthase 2 U53855 at Prostaglandin 12 70 25 1.5 1.5 0.9 1.4 1.2 2.3 1.8 4.2 0.6 > (prostacyclin) synthase) U44750_at NAD 15-OHProstaglandin 420 92 1.2 1.2 13 0.9 0.9 0.5 0.9 0.2 1.0 dehydrogenase AB009372 at Lysophospholipase 92 15 0.8 0.6 0.7 0.4 0.6 0.2 0.3 0.1 03 D89069 f at carbonyl reductase 1 948 129 0.8 0.7 0.8 0.4 0.6 0.2 0.4 0.2 0.9 X95986mRNA#l carbonyl reductase 1 848 103 0.8 0.6 0.7 0.4 0.5 0.2 0.5 0.2 0.8 f at D82071_at prostaglandin D2 synthase 40 4 0.9 0.8 1.0 1.3 1.5 11 1.3 1.3 0.7 2 AV: average; SD: standard deviation ♦Significant gene changes are bolded and underlined.

[0067] Discussion. Gene expression profiling of caecum samples revealed both dose and time dependent changes in the mRNA expression levels of genes reflecting mainly the pharmacological action of the compound on tubulin polymerization, stabilization of microtubules, cell cycle arrest, and induction of cell death. For a description of this action, see Giannakakou P et al, Proc. Natl. Acad. Sci. 97:2904-9 (2000); Rothermel J et al, Semin. Oncol 30(3 Suppl 6):51-5 (2003). The timing of the mRNA expression changes correlated well with the clinical and pathological findings. Beginning at 4 hours, there were time and dose dependent increases in mRNAs of alpha and beta tubulins more pronounced at the high dose. The effect was the strongest on tubulin T beta 15 gene. The expression of the tubulin gamma gene was found up-regulated only at day 7 at both dose levels suggesting its implication mainly in the regeneration of intestinal tract cells rather than the direct pharmacological effect of the compound.

[0068] The main toxicity finding of this pharmacogenomic analysis was the ability of EPO906 to affect the transcript levels of several types of acute phase inflammation related genes: (1) Increased expression levels of mRNA encoding for MHC class I and II molecules, CD74, immunoglobulins, TCR, CD14, OX-45, Fc receptors, complement components, and several other molecules were observed as early as 4h post-treatment. These changes clearly indicate accumulation of immune-competent cell populations into the gastrointestinal tract (caecum) in line with the histopathology finding of epithelial cell necrosis. The expression data suggest accumulation of mainly monocyto-macrophagic cells, B cells followed by T cells (see below), and a relative decrease in the numbers of mastocytes (chymase 1, mast cell proteases 1, 3, 4, 8- 10) and red blood cells (alpha and beta globulins, carbonic anhydrase, ALAD, haeme oxygenases); (2) Low to intermediate changes in the mRNA expression levels of soluble mediators and cell-death related molecules (TNF-R1, IL1 alpha and beta, IL6R, bcl2 related genes, cathepsins, caspases, cytochrome c); (3) Strong and early changes in the mRNA expression level of molecules related to tissue remodelling (coUagens, laminins, actins), vascular events and coagulation (metalloproteinases, calpactins (Annexins), tissue and urinary plasminogen activators, thrombin receptors, factor III); (4) Strong and dose dependent changes of the mRNA expression levels of genes related to the arachidonic acid pathway (FIG. 1). (see, TABLE 7). [0069] Most of the above cited inflammatory gene changes were timely concomitant with the development of diarrhoea and the pathological findings of epithelial cell necrosis and loss of the intestinal barrier function. One can consider that the mRNA expression changes of molecules at so early post-therapeutic time points (before epithelial cell necrosis), would be directly related to the occurrence of diarrhoea. In fact, the intestinal barrier cell turn-over is quite fast and the whole intestinal mucosa is restituted every 3 to 5 days in non-treated animals. EPO906-treated animals exhibited clear dose-dependent changes in the expression of cell cycle-related genes in the intestine. Inhibitory effect on the cell cycle was evident through changes in the levels of mRNA expression of cyclins from 4 hours and on at both doses. Consequently, diarrhoea would occur as a dose-related effect, depending of the severity of cell death/loss and capacity of mucosal cell regeneration.

[0070] In this EXAMPLE, a strong dose-dependent acute phase inflammatory response was initiated as soon as 4h after a 1.75 mg/kg single injection of EPO906. This acute inflammatory reaction was deleterious for the intestinal barrier integrity, impairing cell regeneration and facilitating the development of diarrhoea. The increase of several proteases such as cathepsins, serine proteases, tissue metalloproteinases, and the decrease of caspases and cytochrome c expression clearly indicated increased cell death through autophagy. On the whole, these results suggest that EPO906 induces dose dependent cell death through different mechanisms. [0071] The results of this EXAMPLE show the clear participation of the arachidonic acid pathway in this inflammatory picture. See, FIG. 1. In particular, the time and dose dependent increase of PLA₂ mRNA expression, coupled to the dramatic decrease of the carbonyl reductase and NAD-15-OH prostaglandin dehydrogenase very early at 4 hours, strongly suggest the inhibition of the cyclooxygenase pathway. The upregulation of the cyclooxygenase 2 (cox2) at 4 days, key enzyme for the synthesis of prostaglandins, was too late at 4 days to counteract the inhibitory effect. PLA₂ plays a significant role in inflammation by cleaving arachidonic acid from membrane phospholipids and provide a substrate for cyclo- and lipooxygenases. These enzymes catalyze its conversion into prostanoids and leukotrienes, respectively. Carbonyl reductase and NAD-15-OH prostaglandin dehydrogenase (15-PGDH) are considered to be the key enzymes responsible for the biological inactivation of prostaglandins and related eicosanoids. Strong, dose and time dependent inhibition of these enzymes suggest that accumulation of PLA₂ did not result in synthesis of prostaglandin end-products. Kudo I & Murakami M, Prostaglandins Other Lipid Media. 68-69: 3-58 (2002). The relative inhibition of prostaglandin synthesis (slight increases only for mRNA expression of prostaglandin producing enzymes) was corroborated at the high dose with upregulation of lipooxygenases and downregulation of phospholipase C and lysophospholipase mRNA expressions. TNF can activate mechanistically different caspase-independent cell death pathways resulting in apoptosis, and/or necrosis or autophagy-like morphology. Deiss LP et al, EMBO J 15: 3861-70 (1996).

[0072] The mRNA expression profile from EPO906-treated rats show that soluble mediators such as TNF and IL1, intracellular and extracellular proteases such as cathepsins, annexins and coagulation related tissue factors were upregulated at different time points, mostly concomitant to inflammation and diarrhoea. Beutler B, J Invest Med 43: 227-35 (1995); Dinarello CA, Blood 87: 2095-147 (1996). Increased TNFα secretion was demonstrated in the caecum of high dose EPO906-treated animals on days 5 and 8. Data have shown that intravenous (i.v.) injection of TNF causes acute TNF-Rl-dependent apoptotic detachment of enterocytes at the tip of the villi, as soon as 30 to 90 minutes post-injection. Piguet PF et al, Lab Invest 79:495-500 (1999). More recently, other data emphasized that general caspase inhibition sensitized mice to TNF-R1- mediated toxicity and this sensitization could be completely prevented by PLA₂ inhibition. In this EXAMPLE, the mRNA expression profile from the EPO906-treated rat caecum is consistent with these data indicating the pivotal role of the concomitant adverse interactions of increased PLA /cathepsins, decrease caspases, and TNF-mediated cytotoxicity. Huet G et al, Arthritis Rheum. 36: 772-80 (1993); Cauwels A et al, Nat. Immunol. 4:387-93 (2003). [0073] In addition to their role in TNF and PLA₂-induced cell death, cathepsins (strongly upregulated in caecum of EPO906-treated rats) can degrade extracellular matrix proteins, such as coUagens and laminin, and activate the urokinase plasminogen activator, thereby promoting extracellular proteolytic cascade and detachment of enterocytes. Annexins, also upregulated in the caecum of EPO906-treated rats serve as a profibrinolytic coreceptor for both plasminogen and tissue plasminogen activators and initiate the generation of plasmin. These mRNA expression changes related to extracellular matrix remodelling, and complement/coagulation cascade activation were indicative of vascular and thrombotic events in the intestinal mucosa. Arai H, Prostaglandins Other Lipid Media 68-69: 83-94 (2002). [0074] Although studies with epothilone derivatives indicate a lower recognition by Pgp- 170, the product of the mdrl gene. Lavelle F, Bull Cancer 89: 343-50 (2002), we noted here the increased mRNA level of this gene beginning at 24 hours at the high dose only. The highest expression was at day 4 concomitant with the diarrhoea. This was confirmed by RT-PCR where a dose-dependent increase in the mdrl gene expression in the caecum was found at high dose. At 0.4 mg/kg, however, this effect was only minimal to slight.

[0075] The inflammation and diarrhoea were closely related to the pathological findings of epithelial cell necrosis that leads to the loss of the intestinal electrolyte/water barrier function. The pharmacological action of the compound on cell death was associated with its inflammatory effect. The low level of inflammation at the low dose was probably insufficient for the diarrhoea to occur. Some of the early (4-24 hours) and/or consistently highly expressed genes at the high dose are suitable biomarkers for the development of diarrhoea. These are the proinflammatory molecules PLA₂, proteinases and complement components, the cytokine and cytokine receptor TNF-R1 and IL1 alpha, the neuromediators substance P, alpha2D and alpha2A adrenergic receptors, tachykinin 2, and the water transporter aquaporine 8. These results suggest that a therapeutic intervention is most promising if initiated as soon as possible, before induction of acute phase proteins by EPO906 administration, to counteract the effect of translational processing into the protein version of some of these mRNAs such as TNFα, cathepsin C and PLA₂.

[0076] Recently, literature data suggested that use of TNFα and or PLA₂ inhibitors might reduce or completely prevent development of diarrhoea in animal models of intestinal inflammation, or animals treated with diarrhoea-inducing chemotherapeutic compounds. Dan P et al, Biochemistry 37: 6199-204 (1998); Zhao J et al., AACR, SANGSTAT, (San Francisco, Calif. April 6-10, 2002); Adame Y et al, Gastroenterol. Hepatol 25: 235-9 (2003). In supplement, it was shown that recombinant peptides containing a pro-elafin sequence that inhibits transglutaminase, and an antiflammin sequence that inhibits soluble PLA₂ were the most potent PLA₂ inhibitors. Miele L, J. Clin. Invest. Ill: 19-21 (2003).

[0077] In summary, the occurrence of diarrhoea with EPO906 is concomitant with mRNA expression changes related to inflammatory genes, most notably to the upregulation of PLA , TNF, cathepsins, and their related pathways. EXAMPLE 2

MEASUREMENT OF PHOSPHOLIPASE A₂ ENZYMATIC ACTIVITY IN RAT SERUM

SAMPLES

[0078] Introduction and summary. Gene expression profiling of caecum shows changes in the RNA expression level of secreted phospholipase A₂ IIA (sPLA₂HA). See, EXAMPLE 1. In this EXAMPLE, the enzymatic activity of sPLA₂ was measured in rat serum samples. [0079] Phospholipase A₂ (PLA₂) represents a superfamily of esterases that hydrolyze the sn- 2 ester bond in phospholipids releasing free fatty acids and lysophospholipids. There are several isoforms of PLA₂. "Cytosolic" PLA₂ (cPLA₂ or group IV PLA₂) isoforms are found in the cytosol, have a Ca²⁺ independent activity, have a molecular mass of approximately 85 kDa, and are selective for arachidonylated phospholipids (AA) (Types IVA and B). "Secreted" forms (Types I-III, V, IX, and X IB, IIA and V sPLA₂s) do not show specificity for arachidonic acid (AA) in the sn-2 position of phospholipids and have a Ca²⁺ dependent activity. sPLA₂s are generally localized in cellular granules, have a molecular mass of approximately 15 KDa, and the different isoforms show different pattern of expression among the different tissues. [0080] The ubiquitous nature of PLA s highlights the important role they play in many biological processes, including the generation of proinflammatory lipid mediators such as prostaglandins and leukotrienes, and the regulation of lipid metabolism. Glaser KB, Adv. Pharmacol 32: 31-66 (1995). Type IIA PLA₂ has been postulated to play a principal role in the in inflammation, or in the release of arachidonic acid for metabolism to eicosanoids. Tischfield JA, J Biol. Chem. 272: 17247-17250 (1997).

[0081 ] Dosing group. EPO906 was administered to Lewis rats by ingle dose intravenous administration at 0.4mg/kg and 1.75 mg/kg to groups of 5 animals, which were sacrificed after a recovery period of 4, 24, 48 hrs and 7 days.

[0082] Measurement ofsPLA₂ enzymatic activity. Secreted phospholipase A₂ enzymatic activity assay was performed on rat serum according to the assay procedures described by the kit manufacturers. The activity of sPLA₂ was measured in rat serum 48 hrs and 7 days after a single dose administration of 0.4 mg/kg or 1.75 mg/kg EPO906.

[0083] Quantitative determination of secreted Phospholipase A₂ (sPLA₂) activity in serum was performed with the Correlate-Enzyme Assay sPLA kit (Assay Designs Inc., cat n° 907- 002). In the assay a specific substrate for sPLA₂ is converted into sulfhydryl molecule. The presence of sulfhydryl product was detected colorimetrically using Ellman's reagent, DTNB.

The assay was performed according to the manufacture instructions.

[0084] Sample preparation. Frozen serum samples were thawed at room temperature. All the samples were analyzed undiluted and with a 1 :2 dilution.

[0085] sPLA2 enzymatic activity calibration curve. The calibration concentrations were set according to the kit manufacture instructions. The calibration model (log-log fit) is: log y = A+B*log (x) where y is the optical density; x is the concentration of sPLA (units/ml) in C samples; A is log Y-mteraction of the linear for log(x)=o, B is the curve slope; R is the correlation coefficient for the specific plot with the curve fit. A typical calibration curve is shown in FIG. 5. The software used for calibration and data acquisition was SoftmaxPro® version 3.1.1. [0086] Data analysis. The amount of sPLA activity in the sample was calculated according to calibrators provided with the kit. The values acceptance criteria used was: CV< 20% (coefficient of variation = 100 x SD / mean). The mean value was obtained from the 1:2 diluted and undiluted serum results. Statistical analyses were performed using a single factor ANOVA test.

[0087] Results and conclusions. The measurement of the sPLA activity was performed in two independent assays. In the first assay, sPLA₂ activity was measured in serum of animals treated with 0.4 mg/kg EPO906 and in control animals. In the second assay, the sPLA₂ activity was measured in serum of animals treated with 1.75 mg/kg EPO906 and in control animals. [0088] The difference in the results of sPLA activity observed in the two analyses for control animals are within the variability range of the assay.

[0089] The assay does not discriminate between the enzymatic activity of the sPLA₂ IIA and other sPLA₂ subtypes that can be present in the serum.

[0090] The sPLA₂ activity after 0.4 mg/kg EPO906 treatment: No statistically significant difference in the sPLA enzymatic activity was observed after single dose application of 0.4 mg/kg EPO906 compared to control. See, TABLE 9, FIG. 3.

[0091] The SPLA₂ activity after 1.75 mg/kg EPO906 treatment: A statistical relevant decrease (p>0.001) of sPLA enzymatic activity was observed 48 hrs and 7 days after 1.75 mg/kg EPO906 treatment compared control. See, TABLE 10, FIG. 4. The sPLA₂ mean activity of the treated animals sacrificed after 48 hrs and 7 days is approximately 1.2 and 2.2 fold lower compared to the control animals respectively. TABLE 9 Individual sPLA? activity in serum samples of control and 0.4 mg/kg EPO906 treated animals Treatment Animal SPLA2 activity (units/ml) 1 50.3 2 44.4 Veihcle 3 43.9 4 40.5

17 47.9 0.4 mg/kg: 48 hrs 18 53.5 19 48.9 20 40.7 26 36.2 27 49.5 0.4 mg/kg: 7 days 28 40.0 29 46.5 30 33.0

TABLE 10 Individual sPLA₂ activity in serum samples of control and 1.75 i mg/kg EPO906 treated animals Treatment Animal SPLA2 activity (units/ml) 1 59.9 2 55.9 Veihcle 3 54.3 4 50.4 5 57.2 41 48.5 42 43.3 1.75 mg/kg: 48 hrs 43 41.2 44 41.5 45 46.8 51 31.7 52 31.2 1.75 mg/kg: 7 days 53 19.1 54 29.3 55 16.3

[0092] In summary, gene expression analysis of caecum a dose and time dependent increase of sPLA₂ IIA mRNA was observed after EPO906 treatment, whereas a significant time dependent decrease of sPLA₂ activity could be observed in serum after 1.75 mg/kg EPO906 treatment. EXAMPLE 3

MICROARRAY GENE EXPRESSION ANALYSIS TN RAT CAECUM: COMPARISON

WITH TAXOL

[0093] Introduction and summary. Like EPO906, taxol is a microtubule stabilizing agent that induces cell cycle arrest and induction of apoptosis. Ganansia-Leymarie V et al, Curr. Med. Chem. Anti-Cane. Agents 3: 291-306 (2003). However, contrary to EPO906, diarrhoea is not reported as a side effect in taxol treated patients. To understand the mechanism of diarrhoea induction after EPO906 treatment, gene changes were analyzed in Lewis rat caecum injected with either EPO906 or taxol, to understand the mechanism of diarrhoea after EPO906 treatment, and identify specific biomarkers for efficacy and toxicity. Gene expression profiling was used to define a list of biomarkers by comparing to a low dose (0.4 mg/kg) that did not induced diarrhoea in this model. However, in order to understand the mechanism of diarrhoea induction, a series of assays using a similarly acting compound, like taxol, which did not induce diarrhoea after a single injection of 10 mg/kg, were conducted and the gene expression profile evaluated. [0094] EPO906 at a single i.v. injection of 1.75 mg/kg induced diarrhoea after 4 days while taxol (10 mg/kg single injection, i.v.) did not. Caecum samples were analyzed at 2, 24 and 48 hours after the injections. The pathology findings were diffuse inflammation of the gastrointestinal tract slightly more marked after treatment with EPO906 than with taxol. [0095] Taxol did not induce diarrhoea after a single i.v. injection of 10 mg/kg in the Lewis rat contrary to EPO906 (1.75 mg/kg,). The pathology finding showed diffuse inflammation in intestinal tissues slightly more marked after treatment with EPO906 than with taxol. Caecum samples were analyzed at time 2, 24 and 48 hours after the injections. [0096] Both compounds induce similar pattern of mRNA expression of genes related to tubulins and cell-cycle (pharmacological action). However, most of the changes in terms of number and level of expression of genes was weak or transient with taxol compared to EPO906. The pattern of expression profile of EPO906 was very similar to that observed in EXAMPLE 1. [0097] Also, contrary to EPO906, taxol did not significantly changes the mRNA levels of the proinflammatory cytokines interleukin 1 (IL-1) and the receptor Tumour Necrosis Factor- Receptor 1 (TNF-Rl). Phospholipase A₂ (PLA₂) mRNA expression was strongly upregulated by EPO906 and downregulated by taxol, while the multidrug resistance protein 1 (mdrl) mRNA was upregulated by EPO906 and downregulated by taxol. As a corollary, there were few changes indicative of inflammation, tissue remodelling, and repair with taxol compared to EPO906 effect. [0098] The results of this EXAMPLE show that EPO906 induced far stronger pharmacological and toxicological mRNA expression changes in caecum than taxol. The toxicity of EPO906 seems to be related to its exaggerated pharmacological action on intestinal cell death. The intestine is either less sensitive to taxol, or its intestine tissue exposure was not enough to induce strong pharmacological, and hence toxicological effects. mRNA expression changes related to inflammation, intestinal mucosal cell death and loss of electrolyte/water barrier function were clearly found important to induce diarrhoea in EPO906-treated rats. Several candidate biomarkers for selective and early prediction of diarrhoea could be selected for further evaluation. TNFα and the prostaglandin pathway, in particular PLA₂, need to be further evaluated for their potency of inducing diarrhoea in this model. Finally, we evaluate the anti- diarrhoea effects of inhibitors of PLA₂ and TNFα.

[0099] In summary, similar trends of gene changes were observed with both compounds that could relate to their pharmacological mode of action. However, EPO906 effects were more stringent and last longer at this dose range compared to taxol.

[00100] Test animals and test design. To compare EPO906 with taxol, groups of male Lewis rats received intravenously either a single dose of 1.75 mg/kg EPO906, or 10 mg/kg taxol, or the vehicle only. Samples of caecum were collected at 2, 24, and 48 hours. Forty rats were treated with EPO906, and taxol as comparative compound (TABLE 11). TABLE 11 Group allocation and treatment design Groups Animal No Treatment Time at necropsy 1 1-4 Vehicle non-tumour 2 hours 2 5-8 Vehicle tumour 2 hours 3 9-12 Vehicle tumour 24 hours 4 13-16 Vehicle tumour 48 hours 5 17-20 EPO906; 1.75 mg/kg 2 hours 6 21-24 EPO906; 1.75 mg/kg 24 hours 7 25-28 EPO906; 1.75 mg/kg 48 hours 8 29-32 Taxol; 10 mg/kg 2 hours 9 33-36 Taxol; 10 mg/kg 24 hours 10 37-40 Taxol; 10 mg/kg 48 hours Groups 1-4 and 8-10 received ethanol/Tween-80/saline (5/5/90% v/v) as the vehicle, while groups 5-7 received PEG-300/saline (30/70% v/v).

[00101] Tissue sampling. The animals were sacrificed by exposure to carbon dioxide, followed by necropsy. Portions of the gastrointestinal tract as well all tumours were sampled, fixed in phosphate buffered 10% formalin (v/v), embedded in Paraplast, cut at 4 microns and stained with haematoxylin and eosin. Samples for the gene expression profiling assays were quick-frozen in liquid nitrogen immediately after excision, stored on dry ice and subsequently in a deep-freezer at approximately -80°C until further use (TABLE 12). TABLE 12 Tissue sampling list Sampling Histopathology Genomics EPO906/Taxol Stomach m Duodenum Jejunum m Ileum Caecum Ξ ϋ Colon Rectum Ξ Liver m Sternum with bone marrow Tumour m

[00102] Pathology results. All histologically processed tissues were subjected to microscopic examination. No relevant findings and no differences were noted between non-tumour bearing and tumour-bearing controls.

[00103] In the gastrointestinal tract, there was an increase in mitotic figures especially at 2 hours (mitotic arrest with atypical mitotic figures), accompanied by apoptosis and single cell necrosis beginning at 24 hours. The mucosa appeared diffusely inflamed with a distorted architecture most pronounced after 48 hours. These findings were slightly more marked after treatment with EPO906 than with taxol (TABLE 13). TABLE 13 Summary of the microscopic findings in the large intestine Dose / Compound 1.75 mg/kε EPO906 10 mg/kε Paclitaxel Time after administration 2 h 24 h 48 h 2 h 24 h 48 h Large intestine Increased mitotic figures + (co.,r.) + + + (co.,r.) + (c,r.) + ++ (c.) (rectum) ++ (c.) (rectum) Apoptosis/s.cell necrosis + (co.,r.) + + (c.,r.) + (c,r.) ++ (c.) Diffuse infiamm. (caecum) + + d.: duodenum j. :jejunum c: caecum co.: colon r.: rectum (+): minimal, only one animal affected +: minimal/slight ++: moderate +++: marked These are combined results for groups of 4 rats per time point.

[00104] RNA extraction and purification. RNA extraction and purification was performed as described in EXAMPLE 1.

[00105] Data analysis. Data analyses were performed as described in EXAMPLE 1.

[00106] Gene expression profiling. Pooled data from individual chips were presented in the following TABLES. The significance of a gene expression change was a combination of fold change and results of statistical analysis. This was related to other modulated genes that point to a common biological theme.

[00107] The presence of the tumours did not impact significantly the profile of the gene changes after EPO906 exposure in the non-tumour bearing rats. The gene changes induced by

EPO906 alone in the rat diarrhoea model (one bolus injection of 1.75 mg/kg z.v. in the Lewis rat) are described in EXAMPLE 1 and reproduced in this EXAMPLE. [00108] The pharmacological action of both compounds on tubulins and cell death were signed by the changes in the tubulin genes and markers of cell cycle and apoptosis (TABLE 14 and TABLE 15). TABLE 14 Pharmacological action of EPO906 (1.75 mg/kg) and taxol (10 mg/kg) on tubulins in the caecum < _>f the Lewis rat Control EPO Taxol EPO Taxol EPO Taxol 2 hours 24 hours 48 hours AFFI ID Title AV SD Fold change AB011679_at tubulin, beta 5 111 16 1.2 1.0 L 7 1.3 LI 1.0 rc_AA860030_ _s_at tubulin, beta 5 933 194 1.3 1.2 LI 1.4 LI 1.1 X03369_s_at Rat mRNA for beta- 127 22 1.7 1.4 1 1.3 id 1.0 tubulin T betal5 rc_AA799591. .at ESTs, Highly similar 355 42 1.2 1.3 13 1.2 1.4 1.0 to tubulin T beta 15 rc_AA800948..at ESTs, Highly similar 127 18 11 10 £1 1.6 15 0.8 to tubulin alpha rc_AA892333_ _.at alpha-tubulin 1769 152 1.4 LI 16 1.3 LA 0.9 rc_AI169370_at alpha-tubulin 2841 274 1.3 1.3 LI 1.3 1.3 1.0 V01227_s_at alpha-tubulin 573 197 7.5 1.2 23 1.4 2.0 1.0 Significant gene changes are bolded and underlined.

TABLE 15 Pharmacological action of EPO906 (1.75 mg kg) and taxol (10 mg/kg) on the cell cycle and programmed cell deatr i in the caecum of the Lewis r at Control EPO Taxol EPO Taxol EPO Taxol 2 hours 24 hours 48 hours

AFFI ID Title AV SD Fold chanεe

L13039_s_at calpactin I heavy 1681 236 QA L LI LI 10 1.3 chain

U14647_at caspase 1 277 57 1.1 1.1 07 08 1.0 0.9

U24174_at cyclin-dependent 106 20 1.1 LI 13 1.1 1.3 1.1 kinase inhibitor 1A

U49729_at bcl2-associated X 97 19 1.1 1.0 LI 0.9 1.5 1.0 protein

U59184_at bcl2-associated X 549 98 1.1 1.1 LI 1.1 1.3 1.0 protein

U66470_at cell growth 43 13 0.9 QA 12 LA 15 2.5 regulatory with EF- hand domain

U72350_at B cell lymphoma 2 40 17 13 LI 0.9 0.7 1.6 08 like

U75689_s_at deoxyribonuclease I- 129 10 1.1 0.8 1.1 LA 0.7 like 3

U75689_s_at deoxyribonuclease I- 129 10 1.1 0.8 1.1 Q3 LA 0.7 like 3

X64589_at Cyclin BI 100 31 1.2 1.1 0.8 1.2 01 0.8

X75207_sjιt cyclin DI 230 40 1.1 1.1 1.1 1.0 1.1 QJ.

D16308_at cyclin D2 311 30 1.0 Oil M 1.4 33 1.2 rc_AA899106_at cyclin D2 458 73 1.0 0.9 2.5 1.6 4.5 1.5

Sigmficant gene changes are bolded and underlined.

[00109] In comparison with EPO906, taxol changed significantly the mRNA level of very few genes related to inflammation. In particular, contrary to EPO906, at any time point did taxol increase the mRNA levels of the proinflammatory cytokines or their receptors TNF-Rl and ILl alpha. As a corollary, in contrast to EPO906, there were few changes indicative of vascular events, macrophage priming and repair (TABLE 16). TABLE 16 Selected genes in the inflammatory cascade induced by EPO906 (1.75 mg/kg) and taxol (10 mg/kg) in the caecum of the Lewis rat Control EPO Taxol EPO Taxol EPO Taxol 2 hours 24 hours 48 hours AFFI ID Title AV SD Fold change VASCULAR EVENTS D90404_at cathepsin C 267 38 1.2 1.0 13 1.1 LI 1.2 D90404_g_at cathepsin C 187 56 0.9 0.9 LA 1.0 1.2 0.8 U07619_at Coagulation factor 28 7 0.9 08 10 1.1 1 1.4 III (thromboplastin, tissue factor) M81642_at coagulation factor 469 133 1.1 1.1 13 1.2 LI 0.8 II receptor M23697_at Plasminogen 89 36 0.9 0.8 LA 1.1 12 1.1 activator, tissue M24067_at serine (or cysteine) 37 11 0.9 1.2 19 LA 14 1.6 proteinase inhibitor, member 1 X63434_at Urinary 84 21 1.2 1.0 1.2 1.1 LI 1.0 plasminogen activator, urokinase X71898_at Plasminogen 173 11 1.3 1.3 LA 1.2 1.4 1.0 activator, urokinase receptor

TABLE 16 Selected genes in the mflammatorv cascade induced bv EPO906 (1.75 mg/kg) and taxol (10 mg/kg) in the caecum of the Lewis rat

MACROPHAGE PRIMING rc_AA955600_at Fibronectin 1 42 8 1.0 1.0 1.1 1.1 LA 1.3

U42719_at Complement 308 104 1.2 1.2 LA 1.0 LI 1.1 component 4

M34253_g_at Interferon 615 138 07 1.0 0.8 08 1.0 QJ. regulatory factor 1

X82669complete_seq_at RT1 class lb gene 123 32 1.1 1.0 0.9 1.1 0.9 07

U16025_at RT1 class lb gene, 83 12 1.1 1.1 1.0 0.8 LA 1.1 locus M3

X57523_g_at Transporter 1, ABC 107 64 0.6 0.8 0.9 03 0.6 Q . (ATP binding cassette)

M34253_at Interferon 213 64 06 LA 08 07 1.1 Al regulatory factor 1

D004O3_g_at Interleukin 1 alpha 13 4 18 1.9 1.8 1.2 1.6 0.7

M63122_at Tumour necrosis 241 57 1.0 1.1 LI 1.0 LA 1.1 factor receptor superfamily, member la

AF053312_s_at small inducible 450 227 LA 06 06 M 2.6 0.8 cytokine subfamily A20

REPAIR rc_AI104225_at Laminin chain beta 168 15 1.0 1.1 1.0 1.0 LA 1.2 2

U65656_at Matrix 201 10 1.2 1.1 LI 1.1 LA 0.9 metalloproteinase 2 (72 KDa type TV collagenase)

X70369_s_at collagen, type HI, 925 96 0.9 0.8 1.0 08 11 1.1 alpha 1

Significant gene chanj ges are bolded and underlined.

[00110] Although taxol and EPO906 impacted similar inflammatory genes in the prostaglandin pathway, they were fewer with taxol and, on the whole, downregulated. Of interest was the downregulation of PLA₂, which was contrary to EPO906 effect (TABLE 17). TABLE 17 Genes in the prostaglandin pathway affected bv EPO906 (1 .75 mg/kg) and taxol (10 mg/kg) in the ! caecum of the Lewis rat Control EPO Taxol EPO Taxol EPO Taxol 2 hours 24 hours 48 hours AV SD Fold change AB009372 at Lysophospholipase 64 9 09 0.9 0.6 0.8 0.3 1.0 M99567_at phospholipase C, beta 195 28 LA LA 05 Al 03 0.9 3 X95986mRN carbonyl reductase 1 1096 402 0.9 0.7 0.4 0.6 0.2 0.7 A#l f at D89069_f_at carbonyl reductase 1 1460 389 0.8 0.7 0.4 0.6 0.2 0.6 J03960_at arachidonate 5- 46 7 1.0 1.0 1.2 1.0 10 0.9 lipoxygenase M99567_g_at phospholipase C, beta 116 18 1.1 0.9 Oil <L1 0.5 1.0 3 X51529_at phospholipase A2, 881 586 0.8 0.9 1.1 0.9 IA 07 group IIA (platelets, synovial fluid) U53855_at Prostaglandin 12 130 13 0.9 0.9 1.4 1.1 13 1.0 (prostacyclin) synthase U44750_at NAD-dependent 15- 367 101 1.2 1.1 0.8 0.9 03 0.7 hydroxyprostaglandin dehydrogenase D82071_at prostaglandin D2 44 12 1.0 0.8 1.2 1.3 23 1.1 synthase 2 Significant g ;ene changes are bolded and underlined.

[00111] Discussion. At the gene expression level, both compounds affected the cytoskeleton and programmed cell death. The effect on the cytoskeleton was indicated by the changes in tubulin mRNAs. However, the effects on alpha and beta tubulins mRNAs were more marked and sustained with EPO906 than taxol. In supplement, only EPO906 affected the tubulin beta 15 mRNA.

[00112] The programmed cell death was also affected through changes in proapoptotic molecules and molecules regulating the cell cycle. The caspase 1 was the only caspase altered by both compounds. This gene was downregulated at 24 hours. Both compounds increased the mRNA level of the proinflammatory and proapoptotic molecule calpactin 1, but this effect was stronger and last longer with EPO906 compared to taxol. The effect on the antiapoptotic Bcl2- like molecules also occurred as soon as 24 hours and were more marked after EPO906 treatment.

Cell cycle regulating genes were also changed at this time point, more stringently and sustainably in the EPO906 treated group.

[00113] As described in EXAMPLE 1, EPO906 at 1.75 mg kg single injection in the Lewis rat induced clearly as soon as 24 hours, inflammation and necrosis that involved the coagulation cascade, macrophage priming and repair. In contrast, the very few related genes changes in the inflammatory pathway were downregulated after taxol injection. In particular, after taxol, there were no statistically significant changes in acute phase molecules, and cytokines and their receptors like TNF-Rl and IL-1.

[00114] EPO906-induced inflammation that involved the prostaglandin pathway. Changes in this pathway were described in EXAMPLE 1 and reproduced here. Although, as soon as 2 hours, taxol affected similar genes in this pathway, notably the downregulation of the carbonyl reductases, there was in contrast with EPO906, downregulation of PLA₂ mRNA and no change in prostaglandin synthase and arachidonate 5-lipooxygenase gene expression levels. These are key enzymes in the generation of arachidonic acid. Kudo I & Murakami M, Prostaglandins

Other Lipid Media 68-69: 3-58 (2002).

[00115] We verified that at the gene expression level the ability of microtubule stabilizing agent like taxol to decrease expression of PLA₂ and modulate inflammation. Munns M et al,

Clin. Exp. Pharmacol. Physiol. 26: 230-5 (1999); Moos PJ & Fitzpatrick FA, Proc. Natl Acad.

Sci. USA 95: 3896-901 (1998). This was contrary to EPO906 effect.

[00116] Conclusion. EPO906 induce stronger mRNA expression changes in caecum than taxol. The toxicity of EPO906 seems to be related to its exaggerated pharmacological action on intestinal cell death.

[00117] Inflammation related mRNA expression changes promoting intestinal mucosal cell death and loss of electrolyte/water barrier function were clearly found important to induce diarrhoea in EPO906-treated rats.

[00118] Several biomarkers for selective and early prediction of diarrhoea were identified

(TABLE 18). TABLE 18 Biomarkers of diarrhoea in the caecum of the Lewis rat treated with EPO906 (1.75 mg/kg) All diarrhoea cases EPO906; 1.75 mg/kg. fold change

AFFI m Title Secreted 2_h _4 h 24 h Protein

M63122_at Tumour necrosis factor No 1.0 1.4 1.6 1.4 receptor superfamily, member la rc_AI639488_at ESTs, Highly similar to No 1.1 1.3 1.8 1.6 A42772 mdm2 protein - rat re AA800686 at ESTs, Weakly similar to No 0.9 1.2 1.4 1.5 growth factor receptor bound protein 14 [Rattus norvegicus]

AB011679_at tubulin, beta 5 No 1.2 1.5 1.2 LI X03369_s_at Rat mRNA for beta-tubulin No 1.7 1.4 5JL id T betal5 rc_AA859954_at vacuole Membrane Protein No 1.0 1.1 1.6 LA 1 rc_AI045030_s_at CCAAT/enhancer binding, No 0.9 5.8 4.5 ZA protein (C/EBP) delta

M65149_at CCAAT/enhancer binding, No 1.1 11 2.2 id protein (C/EBP) delta

X60769mRNA_at CCAAT/enhancer binding No 0.9 1.5 LI LΛ protein (C/EBP), beta

X76985_at Latexin No 1.0 1.5 13 LI D78359_at sushi-repeat-containing No 0.8 1.4 1.5 LA protein

M58404_at Thymosin, beta 10 No 0.9 0.8 LA LI re AA875098 at ESTs, Weakly similar to No 0.9 1.1 1.8 1.7 FK506 binding protein 2 (13 kDa) [Rattus norvegicus]

U38253 at eukaryotic translation No 1.0 1.4 1.4 1.5 initiation factor 2B, subunit 3 (gamma, 58kD)

M32062 at Fc receptor, IgG, low No 0.9 0.8 LA LA affinity III

M32062_g_at Fc receptor, IgG, low No 0.8 1.0 1.8 LA affinity III

U02322_s_at neuregulin 1 No 1.1 2.1 1.3 13 AF083269_g_at Actin-related protein No 1.1 0.9 1.2 13 complex lb

X61381cds_s_at interferon induced No 1.0 1.2 1.7 1.8 transmembrane protein 3- like TABLE 18 Biomarkers of diarrhoea in the caecum of the Lewis rat treated with EPO906 (1.75 mg kg) All diarrhoea cases EPO906 i; 1.75 mg/kg, fold change

AFFI ΓD Title Secreted 2 h 4 h 24 h Protein

X83537_at matrix metalloproteinase No 1.1 1.1 1.6 Id 14, membrane-inserted

J00797cds_s_at alpha-tubulin No 1.3 1.3 1.7 LI rc_AI179399_at Collagen, type V, alpha 2 No 1.0 1.4 1.8 Id

M14050_s_at heat shock 70kD protein 5 No 1.0 10 1.3 LA rc_AI113289_s_at protein tyrosine No 1.1 1.1 1.5 id phosphatase, non-receptor type 1

U17565_g_at mini chromosome No 0.9 1.0 Al 06 maintenance deficient 6 (S. cerevisiae) rc_AI639082_s_at mini chromosome No 0.9 1.0 06 06 maintenance deficient 6 {S. cerevisiae) .

K00996mRNA_s_at cytochrome P450, 2b 19 No QJL 0.6 05 1.0

S72505_f_at glutathione S-transferase, No 1.1 0.9 1.7 11 alpha 1

K01932_f_at glutathione S-transferase, No 1.0 0.9 1.4 LA alpha 1 rc_H31313_at selenoprotein P, plasma, 1 Yes 1.1 1.6 13 13

M81855_at P-glycoprotein/multidrug No 1.1 1.0 id 73 resistance 1

M23697_at Plasminogen activator, Yes 0.9 1.6 1.7 LA tissue

D15069_s_at adrenomedullin Yes LI 1.9 1.4 1.2 rc_AI232078_at latent transforming growth No 0.9 1.5 IL 1.4 factor beta binding protein 1

D00403_g_at Interleukin 1 alpha Yes 18 2.2 1.4 1.8 rc_AI169327_g_at tissue inhibitor of Yes 0.9 1.8 73 IA metalloproteinase 1 rc_AI169327_at tissue inhibitor of Yes 1.1 1.4 33 11 metalloproteinase 1

S72594_s_at tissue inhibitor of Yes 0.9 1.0 1.6 LI metalloproteinase 2

X06916_at SI 00 calcium-binding No 0.9 0.8 LI 1.1 protein A4

U15734_at reticulocalbin 2 No 1.0 1.1 15 33

D00753_at Serine protease inhibitor Yes 0.7 1.5 id 10 TABLE 18 Biomarkers of diarrhoea in the caecum of the Lewis rat treated with EPO906 (1.75 mg/kg) All diarrhoea cases EPO906 ; 1.75 mg/kg. fold change

AFFI ID Title Secreted 2 h 4 h 24 h Protein rc_AA800318_at ESTs, Weakly similar to Yes 0.9 1.4 23 LA B26423 serine proteinase inhibitor 2.2

U65656_at matrix metalloproteinase 2 Yes 1.2 1.4 10 13 (72 KDa type IV collagenase)

M15191_s_at Tachykinin (substance P, Yes 0.6 1.0 1.7 LI neurokinin A, neuropeptide K, neuropeptide gamma)

D90404_at cathepsin C Yes 1.2 1.6 1.7 LA

D90404_g_at cathepsin C Yes 0.9 1.3 1.6 LA

AF051895_at Annexin V Yes 1.0 1.3 11 1.3 τc_AA875033_at fibulin 5 Yes 1.1 1.4 LI LA

D88250 at complement component 1, s Yes 0.9 1.3 2.9 1.7 subcomponent rc_AA799803_at ESTs, Weakly similar to Yes 0.9 1.3 2.2 1.5 JC6554 complement subcomponent Cls precursor

X71127_g_at complement component 1, Yes 0.9 1.1 1.8 1.3 q subcomponent, beta polypeptide rc_AA946503_at lipocalin 2 Yes 0.9 1.5 2.0 12. U07619_at Coagulation factor III Yes 0.9 1.3 12 10 (thromboplastin, tissue factor)

M76704_s_at 06-methylguanine-DNA No 1.1 0.9 LI LA mefhyltransferase

D85189_at fatty acid Coenzyme A No 1.0 1.1 1.3 LI ligase, long chain 4

L03294_at Lipoprotein lipase Yes 0.6 1.2 3.1 LI

M15481_g_at insulin-like growth factor 1 Yes 0.9 1.6 2.1 LA rc_AA818593_at phosphatidate No 1.0 1.0 1.3 13 phosphohydrolase type 2a

D88666_at phosphatidylserine-specific Yes 0.8 1.0 LI LA phospholipase Al

D50695_at proteasome (prosome, No 1.0 1.1 1.2 1.3 macropain) 26S subunit, ATPase, 4

D50696 at protease (prosome, No 1.0 1.2 1.4 13 macropain) 26S subunit, ATPase 1 TABLE 18 Biomarkers of diarrhoea in the caecum of the Lewis rat treated with EPO906 (1.75 mg/kg) All diarrhoea cases EPO906 i; 1.75 mε/kε, fold change

AFFI ΓD Title Secreted 2 h 4 h 24 h Protein

M63983_s_at Hypoxanthine No 1.1 1.2 1.0 Id phosphoribosyl transferase

AF009656mRNA_s_at Hypoxanthine No 1.2 1.1 1.5 LA phosphoribosyl transferase rc_AJ237731_s_at Lipoprotein lipase Yes < Z 1.6 4.6 1.8

X76985_at Latexin Yes 1.0 1.5 11 LI rc_AA800844_s_at ESTs, Moderately similar to No 0.9 1.2 10 LI LYOXJtAT Protein-lysine 6-oxidase precursor (Lysyl oxidase) rc_AA859805_at ESTs, Moderately similar to No 1.1 1.3 LI LA LYOX_RAT Protein-lysine 6-oxidase precursor (Lysyl oxidase) rc_AA799755_g_at ESTs, Weakly similar to No 1.6 1.0 1.8 IL carboxypeptidase Z [Rattus norvegicus] rc_AA891842_g_at ESTs, Moderately similar to Yes 0.9 1.0 1.3 13 PODX_RAT Podocalyxin precursor

U69272_at Interleukin 15 Yes 0.8 03 Al 03

AB009999_g_at CDP-diacylglycerol No 1.0 1.2 09 Al synthase (phosphatidate cytidylyltransferase) 1 rc_AI104882_s_at cytosolic epoxide hydrolase No 1.0 1.2 1.0 06

M93297cds_at ornithine aminotransferase No 0.9 1.1 06 06 rc_AA892598_at nucleostemin Unknown 1.0 1.5 1.5 LA

U01145UTR#l_s_at polymeric immunoglobulin No 1.2 08 S3 03 receptor gene

U75928UTR#l_s_at — Unknown 1.0 1.3 LI L rc_AA799448_g_at ESTs Unknown 1.0 07 0.9 1.2 rc_AI638993_s_at ESTs Unknown 0.9 08 0.9 07 rc_AA892986_at ESTs Unknown 1.2 1.0 1.6 LA rc_AA799598_at ESTs Unknown 1.0 1.1 1.6 LI rc_AA893180_at ESTs Unknown 08 1.2 2.0 L2

U02506UTR#l_s_at polymeric immunoglobulin No 1.1 09 08 AA receptor gene rc_AA800790_at ESTs Unknown 1.0 1.6 1.4 LA

D13623_g_at Ribosome-binding protein No 1.0 1.3 1.2 LA p34

X96437 RNA_g_at PRG1 No 1.2 1.1 1.4 LI

Significant changes are bolded and underlined.

[00119] Also, in this EXAMPLE, TNFα secretion in the caecum was demonstrated. [00120] Finally, we evaluate the anti-diarrhoea effects of inhibitors of PLA₂ and TNFα. See, FIG. 6.

EXAMPLE 4

EFFECT OF CELECOXIB TREATMENT ON EPO906-INDUCED DIARRHEA TN THE

LEWIS RAT: MICROARRAY GENE EXPRESSION ANALYSIS IN RAT CAECUM

[00121] Purpose. The purpose of this EXAMPLE was to assess a possible anti-inflammatory and anti-diarrhoeal effect of a combination of the cox2 inhibitor celecoxib with EPO906. The aim was to evaluate the anti-inflammatory and anti-diarrhoea potential of the cyclooxygenase 2 (cox2) inhibitor celecoxib, since an earlier immunohistochemical analysis of Lewis rats treated with EPO906, showed a significant difference in cox2 secretion in correlation with diarrhoea. Celecoxib efficacy was assessed at the doses of 30, or 150 mg/kg given orally for 7 days in combination with EPO906 (1.75 mg/kg i.v., single injection). Caecum samples taken at day 8 were analyzed at the gene expression level.

[00122] The drug combination improved only partly and transiently the clinical diarrhoea readout versus EPO906 alone. However, histopathologically and on the basis of gene expression analysis, there was an increased severity of the lesions of typhlitis in agreement with the inflammatory gene expression profile. The latter indicated considerably more pronounced inflammation with the addition of celecoxib to EPO906 in a dose-dependent fashion. There was a synergistic effect of the drug combination that resulted in a further increase in the mRNA levels of inflammatory cytokine like TNFα and enzymes like PLA₂ to generate inflammatory mediators.

[00123] Treatment. To compare EPO906 alone or in combination with celecoxib, rats received a single intravenous injection of either 1.75 mg/kg EPO906, or multiple oral administration of 150 mg/kg daily (7 days) celecoxib, or a combination of EPO906 and celecoxib at a dose of 30, 50 or 150 mg/kg orally for 7 days, or the vehicle only. Samples of caecum were collected at sacrifice at day 8. [00124] Test animals and test design. Forty eight rats were allocated as follows (TABLE 19). TABLE 19 Group allocation and treatment design Group Animal No EPO906 Λ.v Celecobix fp.o.) Time at necropsy 1 1-8 Vehicle Vehicle 8 days 2 9-16 Vehicle 150 mg/kg 8 days 3 17-24 1.75 mg/kg Vehicle 8 days 4 25-32 1.75 mg/kg 30 mg/kg 8 days 5 33-40 1.75 mg/kg 50 mg/kg 8 days 6 41-48 1.75 mg/kg 150 mg/kg 8 days

[00125] The vehicle for EPO906 was PEG 300 (30%) in NaCl (0.9%) administered intravenously, while for celecoxib, it was the oral administration of a combination of 0.5% methylcellulose and 0.025% Tween 20.

[00126] In vivo examinations. Data collection was performed individually. Clinical signs were assessed using a defined scoring system. Efforts were made to characterize onset and duration of signs observed. Special emphasis was placed on the observation of diarrhoea. Therefore, the animals were kept individually for approximately 3 hours per day from day 2. Diarrhoea was recorded individually with the grading described in EXAMPLE 1. [00127] No diarrhoea was seen in the vehicle or celecoxib only treated groups, while EPO906 induced diarrhoea over days 4-6 post-treatment with a maximum at day 5, as was found in EXAMPLE 1. None of the celecoxib doses (30, 50 or 150 mg/kg) significantly reduced diarrhoea. However, there was a trend for the lower doses of 30 and 50 mg kg to reduce overall diarrhoea and the frequency of grade-3 diarrhoea on days 4 and 5.

[00128] Tissue sampling. The animals were sacrificed by exposure to carbon dioxide, followed by necropsy. Portions of the gastrointestinal tract as well as all tumours were sampled, fixed in phosphate buffered 10% formalin (v/v), embedded in Paraplast, cut at 4 microns and stained with haematoxylin and eosin. Samples for the gene expression profiling assays were quick-frozen in liquid nitrogen immediately after excision, stored on dry ice and subsequently in a deep-freezer at approximately -80°C until further use (TABLE 20). TABLE 20 Tissue sampling list Sampling Histopathologv Genomics EPO906+/- celecoxib Stomach 11 Jejunum m Caecum Liver Ξ Tumour Ξ

[00129] Pathology. All processed tissues for histology were subjected to microscopic examination.

[00130] The treatment with 150 mg/kg/d celecoxib for 7 days induced changes in the stomach only, where slight erosion, focal inflammation and peritonitis were found in single rats. No data on pathological evaluation of the caecum were obtained.

[00131] The treatment with EPO906 alone, or the combination of EPO906 and celecoxib induced changes in the gastro-intestinal tract as well as liver in all animals. In either case, rats displayed minimal to slight increase in mitotic figures (sometimes atypical) and apoptosis/single cell necrosis in the cryptal glands, accompanied by minimal to moderate epithelial regeneration and reactive hypertrophy/hyperplasia. Major microscopic findings are reported in (TABLE 21). TABLE 21 Summary of the microscopic findings in the caecum Group (cage) 1 2 3 4 5 6 EPO906 (mg/kg, i.v.) 0 0 1.75 1.75 1.75 1.75 Celecoxib (mg/kg/day, /?.<?.) 0 150 0 30 50 150 Caecum Increased mitotic figures (glands) + + + + Apoptosis/s.cell necrosis (glands) + + + + Erosion (+) ++ +/++ +/++ Ulceration ++ ++ ++ Diffuse inflammation (mucosa) +/++ +++ ++/-H-+ +++ Focal peritonitis (+) +/++ +/++ +/++ Inflammatory oedema (++) +++ ++/+++ -H-/-H-+ (submucosa) Mucosal regeneration with ++ ++ ++ ++ reactive hypertrophy/hyperplasia Glandular dilatation + + Inc.t + Inc.t + Inc.t ( ): only one animal affected +: minimal/slight ++: moderate +++:marked Inc.t: increased incidence, compared to group 3 (EPO906 alone) I/St: increased incidence/severity, compared to group 3 (EPO906 alone) These are combined results for groups of 8 rats.

[00132] However, only rats freated with the drug combination displayed marked ulceration of the mucosa accompanied by inflammation regardless of the dose of celecoxib.

[00133] In summary, the findings were qualitatively similar, typical of the treatment with

EPO906, however the drug combination increased the severity of the lesions of typhlitis (general inflammation process in the caecum).

[00134] RNA extraction and purification and GeneChip® assay. RNA extraction and purification was performed as described in EXAMPLE 1. GeneChip assays were performed as described in EXAMPLE 1.

[00135] Data analysis. Data analyses were performed as described in EXAMPLE 1. [00136] Gene expression profiling. Pooled data from individual chips were presented in the following TABLES. The significance of a gene expression change was a combination of fold change and results of statistical analysis. This was related to other modulated genes that point to a common biological theme. See, EXAMPLE 1.

[00137] The gene expression profile (GEP) indicated considerably more inflammation with the addition of celecoxib to EPO906, in a dose-dependent fashion, which was consistent with the histopathological picture (TABLE 23). In particular in the arachidonic acid pathway, the extent and degree of gene changes were remarkably enhanced with the combination, notably the changes in cox2 and PLA₂ mRNAs (TABLE 24). Milder but statistically significant similar trend changes existed at 30 mg/kg celecoxib/EPO906 combination.

[00138] Noticeably, profound alterations in the gene expression profile related to the inflammatory pathway existed in the caecum of the rats treated with 150 mg/kg/d celecoxib only. [00139] Discussion. EPO906 which is a novel natural microtubule-targeting agent was previously found to consistently induce diarrhoea in the Lewis rat after a single intravenous injection of 1.75 mg/kg. See. EXAMPLE 1. The occurrence of diarrhoea with EPO906 was concomitant with inflammation and changes in the regulation of a number of genes in the arachidonic acid pathway, notably PLA₂ and to a lesser extent cox2. The purpose of this EXAMPLE was to further explore the effect on diarrhoea by a combination of EPO906 with an anti-inflammatory drug, i.e. celecoxib, which is a cox2 inhibitor. In this EXAMPLE, animals received a single high dose of EPO906, which is associated with diarrhoea, combined with the administration of celecoxib for 7 days. Caecum samples were analyzed at day 8 at necropsy. [00140] The single EPO906 injection at a dose of 1.75 mg/kg induced changes in tubulin mRNAs and molecules that regulate cell cycle and cell death roughly similar to the time course data at day 7 (see, above) except the effect on the beta-tubulin T (beta 15) gene which was absent here. The effect on inflammatory gene changes and alterations in the molecules of the arachidonic acid pathway were qualitatively similar to those observed in a previous EXAMPLE at day 7. See, EXAMPLE 1. However, the degree of changes was on the whole less than previously found, and often time did not reach statistical significance. Interstudy variations and a further recovery of inflammation at day 8 (versus day 7) might account for these discrepancies. [00141] The gene expression profile indicated a consistent inflammation response, both with repeated administration of celecoxib or with the drug combination. The effect was in both cases more pronounced than in the animals treated with EPO906 alone. The effects were reflected in the quality and quantity of genomic changes at all levels of the inflammatory cascade. They were increased in extent and fold changes that frequently reached statistical significance contrary to EPO906-only effect. The nature of the inflammation was qualitatively similar regardless of the treatment groups, i.e. they included platelet activation, vascular events, cellular events (i.e. macrophage activation and matrix degradation), immune response and eventually repair. There was also a qualitatively similar effect on the arachidonic acid pathway that was dramatically enhanced with the combination compared to EPO906 alone. Notably, the dramatic increase in mRNAs of PLA₂ and further downregulation of the carbonyl reductases and NAD-15-OH prostaglandin dehydrogenase indicated synergistic toxicity of the drug combination on the intestine. The pronounced increase in PLA mRNA could be partly explained by the concomitant profound increase in cytokines like IL-1 alpha, beta, and TNFα (reflected by the changes in TNF-Rl mRNA) with the combination, which triggered PLA₂ synthesis. Vadas P et al, Immunol. Lett. 28: 187-93 (1991). Increased PLA enhances arachidonic acid accumulation and synthesis of inflammatory mediators while downregulation of the carbonyl reductases and NAD-15-OH prostaglandin dehydrogenase prevents their deactivation by end-products generation. Although, there was a pronounced increase in prostaglandin synthase mRNAs cox2 with the drug combination, and a slight increase in coxl and PGE2 mRNAs, this was probably too late to prevent mucosal damage. Sigthorsson G et al, Gastroenterology 122: 1913-23 (2002). [00142] A possible role for TNFα, and hence TNF-Rl, in the mechanism of diarrhoea induction was described in EXAMPLE 1. Here, gene expression profile data from caecum of rats treated with the high dose of 1.75 mg/kg EPO906 was consistent with a pivotal role of the concomitant adverse interactions of increased PLA₂/cathepsins, decreased caspases, and TNF- mediated cytotoxicity. Huet G et al, Arthritis Rheum 36: 772-80 (1993); Cauwels A et al, Nat Immunol 4:387-93 (2003). It is remarkable that the change in TNF-Rl was still reaching statistical significance after 8 days. TNF-Rl is normally a partner of acute inflammation being degraded quickly; hence an important role of this cytokine in the EPO906-induced diarrhoea syndrome is likely. At the protein level, increased TNFα secretion was demonstrated in the caecum of 1.75 mg/kg EPO906-treated animals on days 5 and 8.

[00143] In summary, the EPO906 induced inflammatory processes on the histology and gene expression profile in the intestine was not positively influenced by the administration of the anti-inflammatory compound celecoxib which is a cox2 inhibitor. On the reverse, the combination at 30 or 150 mg/kg/d celecoxib for 7 days considerably aggravated the adverse event pathways of inflammation observed with EPO906. This might be due to the synergistic proinflammatory effects of the drug combination to increase further the mRNA levels of inflammatory cytokine like TNFα, interleukin- 1 (IL-1) alpha and beta, and enzymes like PLA₂ to generate inflammatory mediators. The action of a cox2 inhibitor in this process was not efficient to prevent accumulation of PLA₂ mRNA and interaction with TNFα.

TABLE 23 Selected genes in the inflammatory cascade induced bv celecoxib (150 mg/kg, 7 days). EPO906 (1.75 mg/kg) or a combination of EPO906 and celecoxib * Cele EPO EPO + Cele Control 150 1.75 30 150

AFFI fD GENE NAME AV SD FOLD CHANGE

VASCULAR EVENTS

U07619_at coagulation factor 3 28 9 A 3.4 AA 123

M81642_at coagulation factor II receptor 475 123 2.0 1.5 11 LI

M23697_at plasminogen activator, tissue 63 20 12.7 4.6 10.0 11.8

X63434_at plasminogen activator, urokinase 85 23 Al 1.6 33 12

X71898_at Plasminogen activator, urokinase 167 64 10 1.3 LA LA receptor

X52140_at integrin alpha 1 72 21 LI 1.3 1.7 LA

D00913_at intercellular adhesion molecule 1 29 21 iA 1.5 12. id

D87840_at mucosal vascular addressin cell 94 27 11 1.7 11 16 adhesion molecule 1

U08136_at selectin, endothelial cell, ligand 340 53 0.6 06 06 06

S79523_at selectin, lymphocyte 29 17 14 2.0 1.5 12

L23088_at selectin, platelet 26 7 AA 2.1 Al Al

M84488_at vascular cell adhesion molecule 1 11 6 ZA 4.7 Al 63

MACROPHAGE PRIMING rc_AA899935_at platelet-activating factor 44 7 1.5 1.6 LI 1.5 acetylhydrolase alpha 2 subunit (PAF-AH alpha 2)

AF016047_at platelet-activating factor 84 10 07 07 06 06 acetylhydrolase, isoform lb, alphal subunit

D89655_at scavenger receptor class B, 66 7 11 LI 10 L2 member 1

AF071495_s_at scavenger receptor class B, 117 31 11 LI 10 IL member 1

V01216_at orosomucoid 1 15 3 A 0.8 5.0 93

X05834_at fibronectin 1 380 149 A 1.7 A 17

X71127_at complement component 1, q 530 72 id. 2.1 16 12 subcomponent, beta polypeptide

D88250_at complement component 1, s 161 40 116 4.9 Al 10.0 subcomponent

U42719_at complement component 4a 337 124 11 3.5 12

AB003042_at complement component 5, 40 15 AA 1.7 id 43 receptor 1

X61381cds_s_at interferon induced 687 233 IA 23 id transmembrane protein 3-like

D00403_at interleukin 1 alpha 40 18 . 3.4 63 9J>

M98820_at interleukin 1 beta 49 17 18.1 2.2 9.9 23.0

Z22812_at interleukin 1 receptor, type II 47 20 id 1.1 11 Id

X69903_at interleukin 4 receptor 275 82 LA 1.1 1.5 LI TABLE 23 Selected genes in the inflammatory cascade induced by celecoxib (150 mg/kg, 7 days), EPO906 (1.75 mg/kg) or a combination of EPO906 and celecoxib * Cele EPO EPO + Cele Control 150 1.75 30 150

AFFI ID GENE NAME AV SD FOLD CHANGE

MACROPHAGE PRIMING

M58587_at interleukin 6 receptor 51 15 2.2 1.3 1.7 1.9 M92340_at interleukin 6 signal transducer 33 22 iA 2.3 33 £6 U94330_at tumour necrosis factor receptor 49 15 2.0 1.1 1.5 1.7 superfamily, member l ib

M63122_at tumour necrosis factor receptor 226 56 1.6 2.1 1.9 2.3 superfamily, member la rc_AA900380_at tumour necrosis factor receptor 23 13 3.5 1.7 2.5 4.5 superfamily, member lb

X17053mRNA s at small inducible cytokine A2 108 17 10.7 2.2 8.8 10.4

X17053cds_s_at small inducible cytokine A2 63 18 14.1 3.3 13.8 16.2

U06434_at small inducible cytokine A4 12 8 10.1 1.4 6.6 11.8

Y08358_at small inducible cytokine 241 150 3,6 1.2 A 23 subfamily Al l

M32062_at Fc receptor, IgG, low affinity III 89 16 16.5 3.2 2d 13.0 X73371 at low affinity immunoglobulin 40 25 19.6 4.7 12.3 15.8 gamma FC region receptor II precursor

M30691_at Ly6-C antigen gene 102 32 33 1.8 LI 33

Y12502cds_at coagulation factor Xllla 22 15 1.2 id 1.8 1.5

MACROPHAGE PRODUCT AND MATRIX

DEGRADATION rc_AA892775_at Lysozyme 535 326 9.2 3.5 5.8 7.1

S66184_s_at lysyl oxidase 64 26 21 1.6 Zl 83 rc_AI234060_s_at lysyl oxidase 25 15 76 1.3 IA £

J05495_at macrophage galactose N-acetyl- 25 10 11 33 12 Zl galactosamine specific lectin

E13732cds_at macrophage inflammatory 34 8 53 2.2 id 43 protein- 1 alpha receptor gene

S73424_s_at macrophage migration inhibitory 712 117 12 15 23 15 factor rc_AI009801_at macrophage migration inhibitory 119 46 IA i 18 23 factor

M65253_at matrix metalloproteinase 10 190 108 63 0.5 3.7 l U46034_at matrix metalloproteinase 11 24 6 1.5 0.7 1.1 1.7 X83537_at matrix metalloproteinase 14, 237 21 12 2.1 3.8 1 membrane-inserted

U65656_at matrix metalloproteinase 2 (72 210 58 12 2.6 id 53 KDa type IV collagenase)

ABO 10960 s at Matrix metalloproteinase 23 37 18 2.9 23 2.1 2.6 TABLE 23 Selected genes in the inflammatory cascade induced by celecoxib (150 mg/kg, 7 days), EPO906 (1.75 mg/kg) or a combination of EPO906 and celecoxib * Cele EPO EPO + Cele Control 150 1.75 30 150

AFFI ID GENE NAME AV SD FOLD CHANG!

MACROPHAGE PRODUCT AND MATRIX DEGRADATION

X02601_at matrix metalloproteinase 3 16 6 483 1.6 40.1 65.2

Y00497_s__at superoxide dismutase 2 97 29 53 1.7 43 4.5

Z24721_at superoxide dismutase 3 133 54 L2 11 LA 12

U48829_s_at nitric oxide synthase 2, inducible 28 18 33 0.7 1.1 3.1

S71597_s_at nitric oxide synthase 2, inducible 23 7 63 1.1 LA 2

D83661_s_at nitric oxide synthase 2, inducible 17 4 11.0 1.7 33 10.2

D44591_s_at nitric oxide synthase 2, inducible 2 1 83.7 2.3 15.6 82.2

AF006619_s_ at nitric oxide synthase 2, inducible 6 5 12.1 1.6 3.8 12.2

REPAIR

Z78279_at collagen, type 1, alpha 1 353 214 LA 1.6 1.6 11

X70369_s_at collagen, type III, alpha 1 952 533 12 11 23 33

M21354_s_at collagen, type III, alpha 1 377 211 33 14 11 12 rc_AA859757_ at collagen, type V, alpha 1 56 17 10 1.7 LA id

AJ005394_at collagen, type V, alpha 1 306 134 23 10 23 18 rc_AI179399_at collagen, type V, alpha 2 226 70 3.1 2.1 2.8 3.0 re AI232379 at platelet derived growth factor 185 36 33 2.2 receptor, alpha polypeptide a A

Z14120cds_s_at platelet derived growth factor, 66 25 0.5 04 03 AA alpha

D10106_s_at platelet derived growth factor, 138 99 0.6 QA 06 01 alpha

U77697_at platelet/endothelial cell adhesion 49 21 LI 1.6 1.7 L2 molecule

A03913cds_s_at serine (or cysteine) proteinase 27 14 2.7 1.6 id id inhibitor, clade E, member 2

M69246_at serine (or cysteine) proteinase 194 182 11 2.0 2.2 18 inhibitor, clade H, member 1

M24067_at serine (or cysteine) proteinase 35 11 10.7 2.2 6.7 11.4 inhibitor, member 1

D00753_at Serine protease inhibitor 37 10 29.9 6.9 19.7 19.7 S72594_s_at tissue inhibitor of 387 86 4.6 2.6 IA 43 metalloproteinase 2 rc_AA799340_at tissue inhibitor of 296 89 I 1.3 L2 12 metalloproteinase 2

X52498cds_at transforming growth factor, beta 75 30 i 1.7 33 i 1

L09653 s at transforming growth factor, beta 24 13 3.5 2.2 2.7 2.6 receptor II

Significant gene changes are bolded and underlined.

* At day 8, at sacrifice TABLE 24 Genes in the prostaglandin pathway affected bv celecoxib (150 mg/kg, 7 days), EPO906 (1.75 mg/kg) or a combination of EPO906 and celecoxib * Cele EPO EPO + Cele Control 150 1.75 30 150

AFFI ID GENE NAME AV SD FOLD CHANGE

S67722_s_at Prostaglandin- 25 4 11.2 2.1 AA 13.6 endoperoxide synthase

X51529_at phospholipase A2, 787 647 iA 83 93 83 group IIA (platelets, synovial fluid)

L25925_s_at Prostaglandin- 17 10 63 2.2 id 63 endoperoxide synthase

U53855_at prostaglandin 12 101 53 id 1.3 23 15 synthase

U38376_s_at phospholipase A2, 87 31 23 1.1 10 14 group rVA (cytosolic, calcium-dependent) rc_AA998338_s_at phospholipase D2 36 12 10 1.7 12 12 rc_AI104781_at arachidonate 5- 99 38 iA 07 1.5 2.2 lipoxygenase activating protein

X52196cds_at arachidonate 5- 200 30 17 0.8 1.2 10 lipoxygenase activating protein

D28860_s_at prostaglandin E receptor 25 9 2.0 10 id 1.9 4 (subtype EP4)

U03388_s_at Prostaglandin- 94 26 LA 0.9 1.3 LI endoperoxide synthase 1 rc_AI145502_s_at prostaglandin F2 37 12 1.8 1.5 13 1.6 receptor negative regulator

U26595_at prostaglandin F2 179 34 1.7 1.3 LI 1.5 receptor negative regulator

U07798UTR#l_at phospholipase A2, 50 25 0.6 0.7 03 06 group 2C

L15556_at phospholipase C, beta 4 78 20 0.8 06 07 06

J05155_at phospholipase C, 110 51 05 06 03 03 gamma 2

AB000778_s_at phospholipase DI 215 85 0.6 06 04 03

M99567_at phospholipase C, beta 3 259 51 0.4 06 04 Qd

M99567_g_at phospholipase C, beta 3 155 35 0.3 03 04 Qd

U88986_s_at phospholipase DI 47 16 0.3 06 04 Qd TABLE 24 Genes in the prostaglandin pathway affected by celecoxib (150 mg/kg, 7 days), EPO906 (1.75 i mg/kg) or a combination of EPO906 and celecoxib * Cele EPO EPO -l- Cele 1 Control 150 1.75 30 150

AFFI ID GENE NAME AV SD FOLD CHANGE

M20637_s_at phospholipase C, delta 1 93 41 0.3 03 03 Qd

M20637_s_at phospholipase C, delta 1 93 41 0.3 03 03 04

S69383_at arachidonate 12- 191 181 0.3 03 03 03 lipoxygenase

D89069_f_at carbonyl reductase 1 855 429 0.2 04 03 03

X95986mRNA#l _f_at carbonyl reductase 1 636 188 0.2 04 03 03

D89070cds_s_at carbonyl reductase 1 1270 620 0.2 Qd Al 03

U44750_at NAD-dependent 15- 405 93 0.2 03 Qd Qd hydroxyprostaglandin dehydrogenase

AB009372_at lysophospholipase 94 37 02 03 Qd Qd

L06040_s_at arachidonate 12- 131 164 01 Q3 Qd Qd lipoxygenase

Significant gene changes are bold ed and underlined.

* At day 8, at sacrifice

[00144] All references cited herein are incorporated herein by reference in their entirety and for all purposes to the same extent as if each individual publication or patent or patent application was specifically and individually indicated to be incorporated by reference in its entirety for all purposes. In addition, all GenBank accession numbers, Unigene Cluster numbers and protein accession numbers cited herein are incorporated herein by reference in their entirety and for all purposes to the same extent as if each such number was specifically and individually indicated to be incorporated by reference in its entirety for all purposes.

[00145] The present invention is not to be limited in terms of the particular embodiments described in this application, which are intended as single illustrations of individual aspects of the invention. Many modifications and variations of this invention can be made without departing from its spirit and scope, as will be apparent to those skilled in the art. Functionally equivalent methods and apparatus within the scope of the invention, in addition to those enumerated herein, will be apparent to those skilled in the art from the foregoing description and accompanying drawings. Such modifications and variations are intended to fall within the scope of the appended claims. The present invention is to be limited only by the terms of the appended claims, along with the full scope of equivalents to which such claims are entitled.

Claims

CLAIMS We claim:

1. A method for predicting diarrhoea in a subject to whom a microtubule stabilizing agent has been administered, comprising the steps of: (a) obtaining the gene expression profile of the subject, wherein the gene expression profile comprises the gene expression pattern of one or more genes, where the expression patterns of the one or more genes are predictive of the occurrence of diarrhoea in a subject following administration of a microtubule stabilizing agent; (b) determining whether the subject is at risk for developing diarrhoea from the administration of the microtubule stabilizing agent.

2. The method of claim 1 , wherein the subject is a mammal.

3. The method of claim 1 or 2, wherein the subject is a rat or a human.

4. The method of any one of claims 1 to 3, wherein the microtubule stabilizing agent is an epothilone or an analogue.

5. The method of any one of claims 1 to 4, wherein the microtubule stabilizing agent is epothilone B.

6. The method of any one of claims 1 to 5, wherein the gene expression profile of the subject is a gene expression of a sample taken from the subject, wherein the sample is taken from the digestive system.

7. The method of any one of claims 1 to 6, wherein the gene expression profile of the subject is a gene expression of a sample taken from the subject, wherein the sample is taken from the caecum.

8. The method of any one of claims 1 to 7, wherein the expression is an increase in the gene expression of the genes selected from the group consisting of phospholipase A (PLA ); the complement component (Clq); interleukin 1 alpha (ILlα); Tumour Necrosis Factor Receptor 1 (TNF-Rl)); alpha2D adrenergic receptor; alpha2A adrenergic receptor; tachykinin 2; aquaporine 8 and combinations thereof.

9. The method of any one of claims 1 to 7, wherein the expression is an increase in the gene expression of the gene for P-glycoprotein/multidrug resistance 1 (mdrl) gene.

10. The method of any one of claims 1 to 7, wherein the expression is an increase in the gene expression of the genes selected from the group consisting of MHC class I and II molecules; CD74; immunoglobulins; T-cell receptor (TCR); CD 14; OX-45; Fc receptors; complement components and combinations thereof.

11. The method of any one of claims 1 to 7, wherein the expression is a dose dependent change in the gene expression of the gene for cyclooxygenase 2 (cox2).

12. The method of any one of claims 1 to 7, wherein the expression is an early change in the gene expression of genes for tissue remodelling, vascular events and coagulation.

13. The method of claim 12, wherein the genes for tissue remodelling are selected from the group consisting of genes for coUagens, laminins and actins.

14. The method of claim 12, wherein the genes for coagulation are selected from the group consisting of genes for metalloproteinases; calpactins (Annexins); tissue and urinary plasminogen activators; thrombin receptors; Factor III and combinations thereof.

15. The method of any one of claims 1 to 7, wherein the expression is an early increase in the expression of the gene for interleukin 1 alpha; adrenomedullin; or a combination thereof.

16. The method any one of claims 1 to 7, wherein the expression is an early decrease in the expression of the gene for K00996mRNA_s_at; rc_AI237731_s_at (lipoprotein lipase); or a combination thereof.

17. A method for predicting diarrhoea in a subject to whom a microtubule stabilizing agent has been administered, comprising the step of: determining the amount of secreted phospholipase A2 (sPLA₂) protein in the subject, wherein an increase in sPLA₂ is predictive of the occurrence of diarrhoea in a subject following administration of a microtubule stabilizing agent, wherein an increase in the secretion of sPLA₂ protein indicates that the subject is at risk for developing diarrhoea from the administration of the microtubule stabilizing agent.

18. The method of claim 17, wherein the amount of secreted phospholipase A2 (sPLA₂) protein in the subject is determined in a biological sample isolated from said subject.

19. The method of claim 18, wherein the biological sample is taken from the digestive system.

20. The method of claim 19, wherein the biological sample is taken from the caecum.

21. The method of any one of claims 17 to 20, further comprising the step of: administering an anti-diarrhoeal therapy to the subject at risk for developing diarrhoea from the administration of the microtubule stabilizing agent.

22. The method of claim 21, wherein the anti-diarrhoeal therapy is administered to the subject before the induction of acute phase proteins in the subject by the microtubule stabilizing agent.

23. The method of claims 21 or 22, wherein the anti-diarrhoeal therapy is the administration of an inhibitor of phospholipase A₂ (PLA₂).

24. The method of claims 21 or 22, wherein the anti-diarrhoeal therapy is the admimsfration of an inhibitor of tumour necrosis factor alpha (TNFα).

25. The use of epothilone B (EPO906) in the manufacture of a medicament for the treatment of cancer in a patient subpopulation, wherein the members of the patient subpopulation are identified by: (a) obtaining the gene expression profile of a patient, wherein the gene expression profile comprises the gene expression pattern of one or more genes, wherein the expression patterns of the one or more genes are predictive of the occurrence of diarrhoea in a subject following administration of epothilone B; (b) determining whether the patient is at risk for developing diarrhoea from the administration of epothilone B; and (c) identifying those patients who are at low risk as being members of the subpopulation to be treated with epothilone B or those who are at high risk as being members of the subpopulation to be treated with a modified or alternative anti-cancer freatment.

26. The use of claim 25, wherein an increase in the gene expression of the genes selected from the group consisting of phospholipase A₂ (PLA₂); the complement component (Clq); interleukin 1 alpha (ILlα); Tumour Necrosis Factor Receptor 1 (TNF-Rl)); alpha2D adrenergic receptor; alpha2A adrenergic receptor; tachykinin 2; aquaporine 8 and combinations thereof as measured in step (a) is indicative for a patient to be at risk for developing diarrhoea from the administration of epothilone B as determined in step (b).

27. The use of claim 25, wherein an increase in the gene expression of the gene for P- glycoprotein/multidrug resistance 1 (mdrl) gene as measured in step (a) is indicative for a patient to be at risk for developing diarrhoea from the administration of epothilone B as determined in step (b).

28. The use of claim 25, wherein an increase in the gene expression of the genes selected from the group consisting of MHC class I and II molecules; CD74; immunoglobulins; T-cell receptor (TCR); CD 14; OX-45; Fc receptors; complement components and combinations thereof as measured in step (a) is indicative for a patient to be at risk for developing diarrhoea from the administration of epothilone B as determined in step (b).

29. The use of claim 25, wherein the expression as measured in step (a) is a dose dependent change in the gene expression of the gene for cyclooxygenase 2 (cox2) indicative for a patient to be at risk for developing diarrhoea from the administration of epothilone B as determined in step (b).

30. The use of claim 25, wherein the expression as measured in step (a) is an early change in the gene expression of genes for tissue remodelling, vascular events and coagulation indicative for a patient to be at risk for developing diarrhoea from the administration of epothilone B as determined in step (b).

31. The use of claim 30, wherein the genes for tissue remodelling are selected from the group consisting of genes for coUagens, laminins and actins.

32. The use of claim 30, wherein the genes for coagulation are selected from the group consisting of genes for etalloproteinases; calpactins (Annexins); tissue and urinary plasminogen activators; thrombin receptors; Factor III and combinations thereof.

33. The use of claim 25, wherein the expression as measured in step (a) is an early increase in the expression of the gene for interleukin 1 alpha; adrenomedullin; or a combination thereof indicative for a patient to be at risk for developing diarrhoea from the administration of epothilone B as determined in step (b).

34. The use of claim 25, wherein the expression as measured in step (a) is an early decrease in the expression of the gene for 00996mRNA_s_at; rc_AI237731_s_at (lipoprotein lipase); or a combination thereof indicative for a patient to be at risk for developing diarrhoea from the administration of epothilone B as determined in step (b).

35. A kit for use in predicting diarrhoea in a subject to whom a microtubule stabilizing agent has been administered, comprising: (a) a reagent for detecting the gene expression pattern of one or more genes, wherein the one or more genes are biomarkers for predicting diarrhoea in a subject to whom a microtubule stabilizing agent has been administered; and (b) a container for the reagent.

36. The kit of claim 35, wherein the reagent is a gene chip.

37. The kit of claims 35 or 36, further comprising a written product on or in the container describing the use of the biomarker in predicting microtubule stabilizing agent-mediated diarrhoea in subjects.