WO2007071914A1 - Markers - Google Patents

Abstract

The present invention relates to pharmacodynamic markers for CDKIs including the candidate 2,6,9-tri-substituted purine known as seliciclib (roscovitine). The identity of these markers facilitates the convenient identification of seliciclib (roscovitine)-like activity both in vitro and in vivo.

Description

MARKERS

The present invention relates to pharmacodynamic markers for cyclin dependent kinase inhibitors. In particular, the present invention relates to pharmacodynamic markers for the candidate 2,6,9-tri-substituted purine known as seliciclib or roscovitine (CYC 202) and seliciclib- or roscovitine-like compounds. The identity of these markers facilitates the convenient identification of roscovrtine-lϊke activity both in vitro and in vivo.

BACKGROUND TO THE INVENTION

A growing family of cyclin dependent kinase inhibitors (CDKI' s) have been identified. These inhibitors have varying activities against the multiple CDK family members. Generally, these inhibitors bind to the ATP binding pockets of CDKs.

The 2,6,9-tri-substituted purines are becoming a well-studied class of compound showing promise as CDKI' s of use in the treatment of proliferative disorders such as cancers, Leukemias and glomerular nephritis. Fischer P & Lane D (Curr Med Chem (2000), vol. 7, page 1213) provides a detailed review of CDKFs₅ their origins and described activities. In. particular, seliciclib, also referred to herein as roscovitine, has been shown to inhibit CDKl, CDK2, CDK5, CDK 7 and- CDK 9 and to block cell cycle. progression in late Gl/early .S and-in M-phase. This compound, (R)-2-[(l-ethyl- 2-hydroxyethyl)amino]-6-benzylamino-9-isopropylpurine, also known as- R- roscovitine, was first described in WO97/20842 (Meijer L etal) and has since been developed as a promising- candidate anti-cancer agent.

Seliciclib (roscovitine) has greatest in vitro potency against CDK2, CDK7 and CDK9, kinases that are involved in cell cycle progression and transcription (McClue, S.J., et al., International Journal of Cancer, 2002. 102: p. 463-468; Fischer, P.M. and A. Gianella-Borradori, Expert Opin Investig Drugs, 2003. 12(6): p. 955-70). CDK2, in association with its cyclin partners, cyclin E and cyclin A, primarily act in the Gl and S phases of the cell cycle by phosphorylating key substrates such as the retinoblastoma and E2F proteins (Zarkowska, T. and S. Mittnacht, The Journal of Biological

Chemistry, 1997. 272(19): p. 12738-46; Xu, M., et al., 1994. 14(12): p. 8420-31).

Phosphorylation of Rb allows the release of E2F1 and facilitates the entry σf cells into

S phase via the expression of genes required for DNA synthesis. The cyclin A-CDK2

5 mediated phosphorylation of E2F1 in late S phase then switches off the expression of those genes, allowing cells to exit S phase. The CDK-activating kinase, or CAK complex, consists of CDK7, cyclin Η, and MATl. In this context, CDK7 phosphorylates other CDKs, which is an essential step for their activation (Lolli, G., et al., Structure, 2004. 12(11): p. 2067-79). In addition, both CDK7 and CDK9 can

10 phosphorylate the carboxy-terminal domain of the large subunit of RNA polymerase II, thereby activating transcription (Ramanathan, Y., et al.,. Journal of Biological Chemistry, 2001. 276(14): p. 10913-10920). Treatment of human tumour cells in culture with seliciclib results in the inhibition of proliferation and transcription, and induces apoptosis from all phases of the cell cycle (McClue, S.J., et al., International

15 Journal of Cancer, 2002. 102: p. 463-468 , Lam, L.T., et al., Genomebiology.Com, 2001. 2(10): p. 1465-6914; Ljungman, M. and M.T. Paulsen, MoI Pharmacol, 2001. 60(4): p. 785-9; MacCallum, D.E., et al., Cancer Research, 2005. 65(14): p. 5399- 5407) The compound has also been shown to reduce the growth of human tumours in mouse xenograft models (McClue, SJ., et al., International Journal of Cancer, 2002.

20. 102: p. 463-468). As a result of these preclinical findings, seliciclib- is currently being evaluated in clinical trials.

The clinical development of traditional cytotoxic .agents has relied upon the determination of maximal tolerated dose; based on the hypothesis that tumours are 25 proliferating and therefore ought to be more susceptible to cytotoxic agents than non- proliferating normal cells. Seliciclib is one of many compounds resulting from an industry-wide approach to identify more target-oriented therapeutics, including those that act on the cell cycle, cell signalling and angiogenesis. However, it remains important to be able to demonstrate that the target is being modulated in vivo. Accordingly, there is a need to identify "biomarkers" i.e. detectable markers that are associated with drug administration and/or activity.

Such biomarkers can guide dosing and define a maximum effective dose rather than a maximum tolerated dose. In addition, therapeutic biomarkers can enable the presence of a "signature" to be determined to help delineate patient populations that are either susceptible or resistant to a given agent (reviewed in Negm, R. S., M. Verma, and S. Srivastava, Trends in Molecular Medicine, 2002. 8(6): p. 288-293 and Sikora, K., Surrogate endpoints in cancer drug development. Drug Discovery Today, 2002. 7(18): p. 951-956).

SUMMARY OF THE INVENTION

The present invention relates to the identification of protein markers which are pharmacodynamic markers for seliciclib (roscovitine) activity. In particular, the present invention relates to the detection of an altered proteomic profile in treated samples compared to samples from untreated individuals.

Accordingly, in a first aspect there is provided a method of monitoring activity of a CDKI comprising: a) isolating a sample, a "treated sample" from an animal model or human, wherein said animal model or human has been treated with a CDKI; b) determining the^proteomic profile of the treated sample; c) comparing the proteomic profile of the treated sample with a normal sample wherein an altered proteomic profile of the treated sample compared with the normal sample is an indication of CDKI activity.

By a "treated sample" is meant a sample derived from a individual animal or patient who has been treated with a CDKI. The sample is suitably blood, serum or plasma. In detection of proteins in serum and, in particular, in plasma samples of patients, samples are removed and subjected to protein analytical techniques such as SELDI-

TOF MS, as described herein.

Thus, in one embodiment, the proteomic profile is the plasma proteomic profile which is preferably determined by SELDI-TOF MS or 2D PAGE. 5

Suitably, the altered plasma proteomic profile is an alteration in mass peaks.

In one embodiment, the alteration in mass peaks is an alteration in an abundant plasma "protein. Suitably the plasma protein is- any of a 7, 14, 17 or a 28 kDa protein. As 10 described herein, preferred abundant plasma proteins correspond to these proteins. Accordingly, in one embodiment, the 7kDa protein is ApoCI, the 14kDa protein is transthyretin, the 17 kDa protein is ApoAII and the 28 kDa protein is ApoAI.

Detection of an altered proteomic profile may be performed by any one of the methods

15 for protein detection known in the art, particularly by microarray analysis, Western blotting, by analysing protein content of samples using methods such as SELDI-TOF MS as described herein and using further analytical techniques such as 2Dgel electrophoresis. Techniques such as this can be particularly useful for detecting altered expression in the form of alternative post translationally modified forms of a

20. protein; In one embodiment, detection of proteins of these -particular masses forms part of analysing proteomic patterns or profiles in a biological sample and in the analysis of^~ derivation of a proteomic "signature". Such proteomic signature or profile can be used, to indicate exposure to a CDKI, such as seliciclib as welLas monitoring or predicting clinical response.

25

By "7, 14, 17 or 28 kDa" protein is meant a protein that has an apparent mass of approximately 7, 14, 17 or 28 kDa. Reference to a 7, 14, 17 or 28 kDa protein also includes 7, 14, 17 or 28 kDa biomarker clusters. Accordingly, for example, a "7kDa protein" as referred to herein includes proteins having a mass in the range of

30 approximately 6.5 to 7.5 kDa. Suitably the mass is identified by the appearance of the marker on a chip surface in the approximately 7 kDa mass region. The 7 kDa marker referred to herein has been identified to correspond to apolipoprotein CL Accordingly, reference to a "7 kDa protein" as used herein also refers to apolipoprotein CL

Likewise, a "17kDa protein" as referred to herein includes proteins having a mass in the range of approximately 16.5- to 17.5 kDa. Suitably the mass is identified by the- appearance of the marker on a chip surface in the approximately 17 kDa mass region. The 17 kDa marker referred to herein has been identified to correspond to apolipoprotein All.. Accordingly, reference to a "17 kDa protein" as used herein also refers to apolipoprotein AIL

Suitably, the alteration in mass peaks is a mass shift. Such a mass shift can be an increase in molecular weight of the proteins in the treated sample when compared to equivalent proteins in the normal sample. In one embodiment, the alteration is the presence or absence of one or more post translational modifications of an abundant plasma protein in the treated sample compared to the normal sample.

Preferably, the altered expression is due to the binding of a CDKI or a CDKI metabolite to an abundant plasma protein. Where the CDKI is seliciclib, altered expression may be due to the binding of seliciclib or a seliciclib metabolite to the abundant plasma protein. Seliciclib metabolites include PMF30-128.

In particular,_the present application identifies 7 and ITkDa markers which are found in patient serum taken from patients being treated with seliciclib. These proteins are identified to correspond to the proteins apolipoprotein CI and apolipoprotein All, respectively. Accordingly, the detection of the presence of these proteins either by measuring protein or gene expression at the DNA level enables seliciclib administration and/or activity to be monitored.

Suitably, the expression of these proteins is up-regulated after seliciclib (roscovitine) treatment. In addition, the presence of altered post-translationally modified forms of these proteins can be detected, those forms not being detectable or being detectable to a greater or lesser extent prior to seliciclib (roscovitine) treatment.

Accordingly, in another aspect there is provided a method of monitoring activity of a CDEJ comprising: a) isolating a sample, a "treated sample", from an animal model or human, wherein said animal model or human has been treated with a CDKI; b) determining altered expression of at least one of i) a 7 kDa protein or ii) a 17 kDa protein in said treated sample as compared to an untreated control sample as an, indication of CDKI activity.

Suitably, altered expression is an increase in the molecular weight of a 7 or 17 kDa protein. Preferably, the 7 kDa protein is apolipoproteύxCl and the 17 kDa protein is apolipoprotein AIL

Suitably, the CDKI is seliciclib (roscovitine). As used herein the terms "roscovitine" and "R-roscovitine" and "seliciclib" are used, interchangeably, to refer to the compound 2-(R)-( 1 -ethyl-2-hydrσxyethylamino)-6^:;ben2ylamino-9-isopropylpurine, also referred to as CYC202. In its unqualified fornrthe term "roscovitine" is used to include the R-roscovitine, the S enantiomer and racemic mixtures thereof. This compound and its preparation are described in US Patent 6,316,456. Analogues of roscovitine are described, for example, hrWO 03/002565.

In another aspect of the invention there is provided a method of assessing suitable dose levels of roscovitine comprising monitoring the altered expression of at least one abundant plasma protein. Suitably, the abundant plasma protein is a 7, 14, 17 or a 28 kDa protein. In one embodiment, the 7kDa protein is ApoCI, the 14kDa protein is transthyretin, the 17 kDa protein is ApoAII and the 28 kDa protein is ApoAI.

In a further aspect, there is provided a method for identifying a candidate drug having CDKI-like activity comprising administering said candidate drug to an animal model or human and detecting altered expression of at least one of a 7, 14, 17 or a 28 kDa protein in said treated sample as compared to a normal sample as an indication of CDKI activity.

In another aspect there is provided a method for identifying a candidate drug having CDKI-like activity comprising administering said candidate drug to an animal model or human and detecting altered expression olat least one_of i) a 7 kDa protein or ii) a 17 kDa protein in said treated sample as compared to a normal sample as an indication of CDKI activity..

In yet another aspect there is provided the use of at least one of a 7 kDa or 17 kDa protein in the monitoring of activity of a CDKI.

In a further aspect there is provided a kit for assessing the activity of a CDKI comprising antibodies for at least one of a 7 kDa or 17 kDa protein. Suitably, such kits may comprise the antibodies recognising the protein product of a marker identified herein alone or in combination with antibodies directed to another marker identified herein. Further additional markers are described, for example, in PCT/GB2004/001334 and- PCT/GB2004/001337. For altered expression detected by analysis of protein samples, a kitmay comprise a buffer, chip and quality controls (i.e. known positives or negatives) for detection" of a protein such as a 7 kDa or a 17 kDa protein. Suitable buffers and chips are described herein. Suitably, a number of the biomarkers of roscovitine -activity (i.e. a 7, 14, 17 or 28 kDa-. protein as well as protein markers including any of the markers -described in PCI/GB2004/00-1334 and /or PCT/GB2004/001337) may be observed^" in combination. In one embodiment, a combination of the markers may be observed as part of a serum/plasma proteomic profile or "signature".

Antibodies for the markers identified herein may be derived from commercial sources or through techniques which are familiar to those skilled in the art. In one embodiment, and where altered expression manifests itself through the expression of alteration of post translationally-modified forms of a protein biomarker, antibodies specific for those different forms may be used. Suitably, in any method in accordance with the invention, roscovitine is administered to a human over a period of days prior to removing blood samples.

In another aspect there is provided a method for identifying seliciclib (roscovitine) uptake in a patient comprising identifying plasma proteins -associated with a seliciclib metabolite.

In a further aspect there is provided a method for identifying seliciclib (roscovitine) uptake in a patient comprising identifying a shift in the mass of a plasma protein.

In one embodiment of any aspect, the invention may be further developed to use the effect of roscovitine on gene and/or protein expression as a tool in dose titration i.e. fay monitoring the degree and rate of change of proteomic profile, a suitable dose of roscovitine may be determined. Such analysis may further involve correlation of changes of gene expression with the known rate of inhibition of, for example, either CDK2 activity or RB phosphorylation by roscovitine at the same dosage. In this manner, a single measurement of the rate and degree of gene expression may be taken as indicative of further activities of roscovitine.

Response of εr cancer patient to treatment with a particular course of therapy carrbe highly variable. Different individuals may exhibit variances in- drug metabolism for example, due to enzyme polymorphisms. For example, a patient may be sensitive to treatment with a particular therapy and therefore exhibit reduced tumour burden or improved symptoms. Alternatively, a patient may be resistant to treatment and show no or little improvement in response to a particular therapy. Detecting the presence of a 7 or 17 kDa protein as described herein whose expression is modified by a CDKI such as roscovitine may also be useful in the prediction and monitoring of a response to treatment with a CDKI. Preferably, where roscovitine is administered to a human, the effective concentration of roscovitine administered to a cell is greater than 5 micromolar and, more preferably greater than 10 micromolar.

Suitably, where roscovitine is administered to a human, treatment with the drug is for a period of days prior to removing blood-samples for analysis.

In one embodiment, where roscovitine is administered to a cell, the effective concentration of roscovitine is preferably upto 75 micromolar.

In a preferred embodiment of the invention roscovitine is administered to a mammal or a human, more preferably a human. When performed on an animal model, the invention is preferably performed on a tumour model such as HT29 or A549 xenograft mouse model.

BRIEF DESCRIPTION OF THE FIGURES AND TABLES

FIGURE LEGENDS

Figure 1: Seliciclib-induced changes in plasma proteomic profiles. (A) Composite Biomarker Wizard plot from analysis on GMlO and QlO .chips of plasma from 48 patients either pre (G) or post (o) seliciclib -treatment. Only the mass region between 5 kDa and 30 TcDa is shown. (B) Analysis of the CMlO array spectra between 6.4^~kDa and 7.2 kDa, displayed in gel view format. Spectra from patients before and after treatment are shown listed by patient number (C) Biomarker Wizard plot summarising the data for all 48 patients from the 7kDa region on the CMlO chiμ surface. Included are representative spectra from the CMlO chip of plasma taken from two patients pre and post treatment. Candidate biomarker peaks are indicated by arrows in B and C.

Figure 2: Identification of biomarkers. (A) Plasma samples from patient 113 pre and post seliciclib treatment were analysed by two dimensional gel electrophoresis. The two clusters of spots that showed differences between the pre and post treatment samples are highlighted in boxes. (B) An expanded view of the lower boxed area is shown for plasma samples from two patients. The protein spots are ~14 kDa in size.

The additional spot, which appears more abundant following treatment in both patients, is highlighted with arrows. (C) Comparison between the original SELDI-TOF MS spectra for the 14 kDa marker and the proteins extracted from spots 1 and 2 from the polyacrylamide gels. Protein samples eluted from the gel spots were applied to an

NP20 chip and the profile was compared with the original SELDI-TOF MS spectra.

The extracted proteins aligned with peaks on the original spectra (indicated by arrows).

It should be noted that "the two additional peaks in the day 10 sample on the QlO chip at ~15 kDa were due to haemoglobin contamination of this sample and do not represent a biomarker. (D) An expanded view of the upper boxed area in A is shown for plasma samples from two patients. The spots of interest lie around 28 kDa.

Figure 3: Confirmation of protein identity by immunocapture. Spectra from preactivated chip surfaces bound with antibody to either Transthyretin (A) or

Apolipoprotein Cl (B). For both experiments, spectra are shown for patient 113 before and after treatment on the QlO or CMlO chip and on the preactivated RSlOO chip. In both cases, the biomarker peaks that appeared on the CMlO and QlO chip surfaces in response to treatment with seliciclib aligned with the peaks on the respective antibody chips. Marker peaks are highlighted with^" their molecular weight.

Figure 4: Use of mice as a model system to study seliciclib-induced biomarkers.

(A) Samples were obtained from a mouse xenograft experiment, in which mice were treated with 150 mg/kg seliciclib bid for 5 days. The plasma was run on a CMlO chip and the mass region around the ApoCI marker was compared to a sample from patient 113 in the Phase Ib trial. The equivalent biomarker peak in human and mouse samples is indicated by arrows. (B) Mice were treated with 400mg/kg seliciclib or D7- seliciclib bid for 4 days. Mouse plasma was analysed on a CMlO chip and the mass region around the ApoCI marker was compared between the different treatments. Figure 5: Primary metabolism of seliciclib. The principle phase I metabolite (named PMF30-128) of seliciclib is formed through the sequential oxidation of the primary alcohol group to the corresponding carboxylic acid via a reactive aldehyde intermediate.

DETAILED DESCRIPTION OF THE INVENTION

The practice of the present invention will employ, unless otherwise indicated, conventional techniques of chemistry, molecular biology, cell biology, microbiology, recombinant DNA and immunology, which are within the capabilities of a person of ordinary skill in the art. Such techniques are explained in the literature. See, for example, J. Sambrook, E. F. Fritsch, and T. Maniatis, 1989, Molecular Cloning: A Laboratory Manual, Second Edition, Books 1-3, Cold Spring Harbor Laboratory Press; Ausubel, F. M. et al. (1995 and periodic supplements; Current Protocols in Molecular Biology, ch. 9, 13, and 16, John Wiley & Sons, New York, N. Y.); B. Roe, J. Crabtree, and A. Kahn, 1996, DNA Isolation and Sequencing: Essential Techniques, John Wiley-& Sons; J. M. Polak andJames O'D. McGee, 1990, In Situ Hybridization: Principles and Practice; Oxford University Press; M. J. Gait (Editor), 1984, Oligonucleotide Synthesis: A Practical Approach, IRL Press; and, D. M. J. Lilley and J. E. Dahlberg, 1992, Methods ofEnzymology: DNA Structure Part Ai Synthesis and Physical Analysis of DNA Methods in Enzymology, Academic Press. Each of these general texts is herein incorporated by reference.

By "CDKI" is meant an inhibitor of CDK activity. Seliciclib (roscovitine) is just one of a number of compounds known to be inhibitors of CDK activity.

By "roscovitine activity" or "roscovitine-like activity" is meant an activity exhibited by roscovitine. For example, roscovitine-like means capable of inhibiting cell cycle progression in late Gl/early S or M phase. Preferably, said inhibition of cell cycle progression is through inhibiting CDKs including CDKl, CDK2, CDK5, CDK7 and CDK9. A study of roscovitine activity is reported in McClue et al. Int. J. Cancer, 2002, 102, 463-468.

The term "marker" or "biomarker" of roscovitine activity is used herein to refer to a gene or protein whose expression in a sample derived from a cell or mammal is altered or modulated, for example-, up or down regulated, in response to treatment with roscovitine. Where the biomarker is a protein, modulation or alteration of expression encompasses modulation through different post translational modifications ..

Also used herein is the term "biomarker cluster" which means a group of distinct protein forms having a similar mass, when separated by SELDI-TOF MS. Biomarker clusters are described in the Examples section herein.

For protein analysis, a sample can be a blood, plasma or serum sample.

By "altered expression" is meant an increase, decrease or otherwise modified level or pattern of expression in a sample derived from a treated cell when compared to an untreated, normal control sample.

The term "expression" refers to the transcription of a gene's DNA template to produce the .corresponding mRKA and translation of this mRNA to produceJhe corresponding gene product (i.e., a peptide, polypeptide, or protein) as well as the "expression" of a protein including its one or more forms that may have been modified post translation.

Post translational modifications are covalent processing events that change the properties of a protein by proteolytic cleavage or by addition of a modifying group to one or more amino acids. Common post translational modifications include phosphorylation, acetylation, methylation, acylation, glycosylation, GPI anchor, ubiquitination and so forth. A review of such modifications and methods for detection may be found in Mann et al. Nature Biotechnology March 2003, Vol. 21, pages 255- 261. Protein post translational modifications may also include mass changes which result from adduct formation between a compound administered to a patient or a metabolite of that compound.

By "polynucleotide" or "polypeptide" is meant the DNA and protein sequences encoding the proteins disclosed herein whose expression is modified in response to roscovitine. The terms also include close variants of those sequences, where the variant possesses the same biological- activity as the reference sequence. Such variant sequences include "alleles" (variant sequences found at the_ same genetic locus in the same or closely-related species), "homologues" (a gene related to a second gene by descent from a common ancestral DNA sequence, ^nd separated by either speciation

("ortholog") or genetic duplication ("paralog")), so long as such variants retain the same biological activity as the reference sequence(s) disclosed herein.

The invention is also intended to include detection of genes having silent polymorphisms and conservative substitutions in the polynucleotides and polypeptides encoding the proteins disclosed herein, so long as such variants retain the same biological activity as the reference sequeπce(-s) as diselose&herein.

Measuring altered expression ofprotein markers of CDKI activity

Protein expression may be determined, using a number of different techniques.

Altered gene or protein expression may be detected by measuring the polypeptides encoded by the gene markers of roscovitine activity. This may be achieved by using molecules^" which bind to the polypeptides encoded by any one of the genes or proteins identified herein as a marker of roscovitine activity. Suitable molecules/agents which bind either directly or indirectly to the polypeptides in order to detect the presence of the protein include naturally occurring molecules such as peptides and proteins, for example antibodies, or they may be synthetic molecules. Methods for production of antibodies are known by those skilled in the art. If polyclonal antibodies are desired, a selected mammal (e.g., mouse, rabbit, goat, horse, etc.) is immunised with an immunogenic polypeptide bearing an epitope(s) from a polypeptide. Serum from the immunised animal is collected and treated according to known procedures. If serum containing polyclonal antibodies to an epitope from a polypeptide contains antibodies to other- antigens, the polyclonal antibodies can be purified by immunoaffmity chromatography. Techniques forproducing and processing polyclonal antisera are known in the- art. In order to generate a larger immunogenic response, polypeptides or fragments thereof may be haptenised to -another polypeptide for use as immunogens in animals or humans.

Monoclonal antibodies directed against epitopes in polypeptides can also be readily produced by one skilled in the art. The general methodology for making monoclonal antibodies by hybridomas is well known. Immortal antibody-producing cell lines can be created by cell fusion, and also by other techniques such as direct transformation of B lymphocytes with oncogenic DNA, or transfection with Epstein-Barr virus. Panels of monoclonal antibodies produced against epitopes in the polypeptides of the invention can be screened for various properties; i.e., for isotype and epitope affinity.

An alternative technique involves screening phage display libraries where, for example the phage express scFv fragments on the surface of their coat with a large variety of complementarity determining regions (CDRs). This technique is well known in the art

For the purposes of this invention, the term "antibody", unless specified to the contrary, includes fragments of whole antibodies which retain their binding activity for a target antigen. Such fragments include Fv, F(ab') and F(ab')₂ fragments, as well as single chain antibodies (scFv). Furthermore, the antibodies and fragments thereof may be humanised antibodies, for example as described in EP- A-239400.

Standard laboratory techniques such as immunoblotting as described above can be used to detect altered levels of markers of roscovitine activity, as compared with untreated cells in the same cell population. Gene or protein expression may also be determined by detecting changes in post- translational processing of polypeptides or post-transcriptional modification of nucleic acids. For example, differential phosphorylation of polypeptides, the cleavage of polypeptides or alternative splicing of RNA, and the like may be measured. Levels of expression of gene products such as polypeptides, as well as their post-translational modification, may be detected using proprietary protein assays or techniques such as

2D polyacrylamide gel electrophoresis.

Antibodies may be used in detecting markers of roscovitine activity identified herein in biological samples by a method which comprises: (a) providing an antibody of the invention; (b) incubating a biological sample with said antibody under conditions which allow for the formation of an antibody-antigen complex; and (c) determining whether antibody-antigen complex comprising said antibody is formed.

Antibodies that specifically bind to protein markers of roscovitine activity can be used in diagnostic methods and kits that are well known to those of ordinary skill in the art to detect or quantify the markers of roscovitine activity proteins in a body fluid or tissue. Results from these tests can be used to diagnose or predict the occurrence or recurrence of a cancer and other cell cycle progression-mediated diseases or to assess the effectiveness of drug dosage and treatment.

Antibodies can be assayedJbr irnmunospecific binding by any method known in the art. The immunoassays- which can be used include but are not limited to competitive and^" non-competitive assay systems using¹- techniques such as western blots, immunohistochemistry, radioimmunoassays, ELISA, sandwich immunoassays, imrnunoprecipitation assays, precipitin reactions, gel diffusion precipitin reactions, immunodiffusion assays, agglutination assays, complement-fixation assays, immunoradiometric assays, fluorescent immunoassays and protein A immunoassays. Such assays are routine in the art (see, for example, Ausubel et ah, eds, 1994, Current Protocols in Molecular Biology, Vol. 1, John Wiley & Sons, Inc., New York, which is incorporated by reference herein in its entirety). Antibodies for use in the invention may be bound to a solid support and/or packaged into kits in a suitable container along with suitable reagents, controls, instructions and the like.

Other methods include 2D-P AGE as well as matrix-assisted laser desorption ionization time of flight mass spectrometry (MALDI-TOF MS). In MALD-I-TOF analysis, proteins in a complex mixture are affixed to a solid metallic matrix, desorbed with a pulsed laser beam to generate gas-phase ions that traverse a field-free flight tube, and are then separated according to their mass-dependent velocities. Individual proteins and peptides can be identified through the use of informatics tools to search protein and peptide sequence databases. Surface-enhanced laser desorption/ionisation time of flight MS (SELDI-TOF MS) is an affinity-based MS method in which proteins are selectively adsorbed to a chemically modified solid surface, impurities are removed by washing, an energy-absorbing matrix is applied, and the proteins are. identified by laser desorption mass analysis.

In order to identify protein biomarkers, SELDI-TOF-MS can be used for the detection of the appearance/loss of either intact proteins or fragments of specific proteins. In addition SELDI-TOF-MS can also be used for detection of post translational modifications of proteins due to the difference in mass caused by the addition/removal of chemical groups. Thus phosphorylation of a single residue will cause a mass shift of 80 Da due to the phosphate group. A data base of molecular weights that can be attributed to post-translational modifications is freely accessible on the internet Qittpi//www.abrf.org/index.cfm/dm.home?avgmass=alD. Moreover specific polypeptides can be captured by affinity-based approaches using SELDI-TOF-MS by employing antibodies that specifically recognise a post-translationally modified form of the protein, or that can recognise all forms of the protein equally well.

SELDI-TOF MS combines protein purification chromatography with mass spectrometry in order to enable the analysis of proteomic patterns in complex biological samples. SELDI-TOF MS is being widely used to generate serum proteomic "signatures", for the purposes of diagnosis and early detection of cancer, in particular ovarian and prostate cancer (Petricoin, E.F., III, et al., The Lancet, 2002. 359: p. 572-

577; Rai, AJ., et al., Arch Pathol Lab Med, 2002. 126: p. 1518-1526; Qu, Y., et al.,

Clin Chem, 2002. 48(10): p. 1835-1843; Petricoin, E.F., III, et al., J Natl Cancer Inst, 2002. 94(20): p. 1576-1578 and Zhang, Z., et al., Cancer Research, 2004. 64(16): p.

5882-5890).. In the latter case, a panel of SELDI markers has been shown to confer improved sensitivity and specificity over the current prostate specific antigen test

(Petricoin, E.F., III, et al., J Natl Cancer Inst, 2002. 94(20): p. 1576-1578). In addition. to its use in diagnosis, SELDI-TQF MS has been used to monitor changes in plasma proteomic profiles following drug treatment. Markers have been discovered that were induced by the treatment of familial adenomatous polyposis patients with celecoxib, including a ~17 kDa marker that was a strong discriminator between responding and non-responding patients (Xiao, Z., et al., Serum Proteomic Profiles Suggest Celecoxib- Modulated Targets and Response Predictors. Cancer Res, 2004. 64(8): p. 2904-2909). A second study observed one marker at ~2.8 kDa that was induced in breast cancer patients following paclitaxel treatment (Pusztai, L., et al., Pharmacoproteomic analysis of prechemotherapy and postchemotherapy plasma samples from patients receiving neoadjuvant or adjuvant chemotherapy for breast carcinoma. Cancer, 2004. 100: p. 1814-1822). Work is ongoing to identify these protein markers, which may reveal important information about the mechanism of action of these drugs.

SELDl-TOF-MS can be utilised to analyse and monitor serum proteomic patterns in patients before an&after treatment with a CKDI such as roscovitine. Proteins such as the 7 and 17 kDa markers identified herein can be detected as an indication of roscovitine administration and/or activity. Such serum proteomic patterns may be applied to assess drug efficacy. Generation of mass spectra from serum samples requires only a small serum sample and can provide quick results. Advantages of such proteomic markers are described for example by Petricoin et al., The Lancet, vol. 359, 2002, 572-577. ARRAYS

Array technology and the various techniques and applications associated with it is described generally in numerous textbooks and documents. These include Lemieux et al, 1998, Molecular Breeding 4:277-289; Schena and Davis. Parallel Analysis with Biological Chips, in PCR Methods Manual (eds. M. Innis, D. Gelfand, J. Sninsky); Schena and Davis, 1999, Genes, Genomes and Chips. In DNA Microarrays: A Practical Approach (ed. M. Schena), Oxford University Press, Oxford, UK, 1999); The Chipping Forecast- (Nature Genetics special issue; January 1999 Supplement); Mark Schena (Ed.), Microarray Biochip Technology, (Eaton Publishing Company); Cortes, 2000, The Scientist 14(17):25; Gwynne and Page, Microarray analysis: the next revolution in molecular biology, Science, 1999, August 6; Eakins and Chu, 1999, Trends in Biotechnology, 17:217-218, and also at various world wide web sites.

Array technology overcomes the disadvantages with traditional methods in molecular biology, which generally work on a "one gene in one experiment" basis, resulting in low throughput and the inability to appreciate the "whole picture" of gene function. Currently, the major applications for array technology include the identification of sequence (gene / gene mutation) and the determination of expression level (abundance) of genes. Gene expression profiling may make use of array technology, optionally in combination with proteomics techniques (Celis et al, 2000, FEBS Lett, 480(l):2-16; Lockhart and Winzeler, 2000, Nature 405(6788): 827-836; Khan et- al, 1999, 20(2):223-9). Other applications of array technology are also known in the art; for example, gene discovery, cancer research (Marx, 2000, Science 289: 1670-1672; Scherf et alef al, 2000, Nat Genet 24(3):236-44; Ross et al, 2000, Nat Genet 2000, 24(3):227-35), SNP analysis (Wang et al, 1998, Science 280(5366):1077-82), drug discovery, pharmacogenomics, disease diagnosis (for example, utilising microfluidics devices: Chemical & Engineering News, February 22, 1999, 77(8):27-36), toxicology (Rockett and Dix (2000), Xenobiotica 30(2): 155-77; Afshari et al, 1999, Cancer Res 59(19):4759-60) and toxicogenomics (a hybrid of functional genomics and molecular toxicology). The goal of toxicogenomics is to find correlations between toxic responses to toxicants and changes in the genetic profiles of the objects exposed to such toxicants (Nuwaysir et ah, 1999, Molecular Carcinogenesis 24:153-159).

In the context of the present invention, array technology can be used, for example, in the analysis of the expression of one or more of the protein markers of roscovitine activity identified hereύxJn one embodiment, array technology may be used to assay the effect of a candidate compound on a number of the markers of roscovitine activity identified herein simultaneously.

In general, any library or group of samples may be arranged in an orderly manner into an array, by spatially separating the members of the library or group. Examples of suitable libraries for arraying include nucleic acid libraries (including DNA, cDNA, oligonucleotide, etc. libraries), peptide, polypeptide and protein libraries, as well as libraries comprising any molecules, such as ligand libraries, among others. Accordingly, where reference is made to a "library" in this document, unless the context dictates otherwise, such reference should be taken to include reference to a library in the form of an array. In the context of the present invention, a 'library" may include a sample of markers of roscovitine activity as identified herein.

The samples (e.g.-, members of a library) are generally fixed or immobilised onto a solid phase, preferably a solid substrate, to limit diffusion and admixing of the samples. In a preferred embodiment, libraries of DNA binding ligands may be prepared. In particular, the libraries may be immobilised to a substantially planar solid phase, including membranes and non-porous substrates such as plastic and glass. Furthermore, the samples are preferably arranged in such a way that indexing (i.e., reference or access to a particular sample) is facilitated. Typically the samples are applied as spots in a grid formation. Common assay systems may be adapted for this purpose. For example, an array may be immobilised on the surface of a microplate, either with multiple samples in a well, or with a single sample in each well. Furthermore, the solid substrate may be a membrane, such as a nitrocellulose or nylon membrane (for example, membranes used in blotting experiments). Alternative substrates include glass, or silica based substrates. Thus, the samples are immobilised by any suitable method known in the art, for example, by charge interactions, or by chemical coupling to the walls or bottom of the wells, or the surface of the membrane. Other means of arranging and fixing may be used, for example, pipetting, drop-touch, piezoelectric means, ink-jet and bubblejet technology, electrostatic application, etc. In the case of silicon-based chips, photolithography may be utilised to arrange and fix the samples on the chip.

The samples may be arranged by being "spotted" onto the solid substrate; this may be done by hand or by making use of robotics to deposit the sample. In- general, arrays may be described as macroarrays or microarrays, the difference being the size of the sample spots. Macroarrays typically contain sample spot sizes of about 300 microns or larger and may be easily imaged by existing gel and blot scanners. The sample spot sizes in microarrays are typically less than 200 microns in diameter and these arrays usually contain thousands of spots. Thus, microarrays may require specialized robotics and imaging equipment, which may need to be custom made. Instrumentation is described generally in a review by Cortese, 2000, The Scientist 14(11):26.

Techniques for producing immobilised libraries of DNA molecules have been described in the art Generally, most prior art methods described how to synthesise single-stranded nucleic acid molecule libraries, using for example masking techniques to build up various permutations of sequences at the various discrete positions on the solid substrate. U.S. Patent No. 5,837,832, the_ contents of which- are incorporated herein by reference, describes an improved method, for producing DNA arrays immobilised to silicon substrates based on very Jarge scale integration technology. In particular, U.S. Patent No. 5,837,832 describes a strategy called "tiling" to synthesize specific "sets of probes at spatially-defined locations on a substrate which may be used to produced the immobilised DNA libraries of the present invention. U.S. Patent No. 5,837,832 also provides references for earlier techniques that may also be used.

Arrays of peptides (or peptidomimetics) may also be synthesised on a surface in a manner that places each distinct library member {e.g., unique peptide sequence) at a discrete, predefined location in the array. The identity of each library member is determined by its spatial location in the array. The locations in the array where binding interactions between a predetermined molecule (e.g., a target or probe) and reactive library members occur is determined, thereby identifying the sequences of the. reactive library members on the basis of spatial location. These methods are described in U.S. Patent No. 5,143,854; WO 90/15070 and WO 92/10092; Fodor et al, 1991,

Science 251 :767; Dower and Fodor, 1991, Ann. Rep. Med. Chem. 26:271.

To aid detection, targets andprobes may be labelled with any readily detectable reporter, for example, a fluorescent, bioluminescent, phosphorescent, radioactive, etc reporter. Such reporters, their detection, coupling to targets/probes, etc are discussed elsewhere in this document. Labelling of probes and targets is also disclosed in Shalon et ah, 1996, Genome Res 6(7):639-45.

Data analysis is also an important part of an experiment involving arrays. The raw data from a microarray experiment typically are images, which need to be transformed into gene expression matrices - tables where rows represent for example genes, columns represent for example various samples such as tissues or experimental conditions, and numbers in each cell for example characterize the expression level of the particular gene in the particular sample. These matrices have to be analyzed further, if any knowledge about the- underlying biological processes is to be. extracted. Methods of data analysis (including supervised and unsupervised data analysis as well^" as bioinformatics approaches) are disclosed in Brazma and ViIo J, 2000, FEBS Lett 480(l):17-74.

As disclosed above, proteins, polypeptides, etc may also be immobilised in arrays. For example, antibodies have been used in microarray analysis of the proteome using protein chips (Borrebaeck CA, 2000, Immunol Today 21(8):379-82). Polypeptide arrays are reviewed in, for example, MacBeath and Schreiber, 2000, Science, 289(5485):1760-1763.

DIAGNOSTICS AND PROGNOSTICS The invention also includes use of the markers of roscovitine activity, antibodies to those proteins, and compositions comprising those proteins and/or their antibodies in diagnosis or prognosis of diseases characterized by proliferative activity, particularly in individuals being treated with roscovitine. As used herein, the term "prognostic method" means a method that enables a prediction regarding the progression of a disease of a human or animal diagnosed with the disease, in -particular, cancer. In particular, cancers of interest with respect to roscovitine treatment include breast, lung, gastric, head and neck, colorectal, renal, pancreatic, uterine, hepatic, bladder, endometrial and prostate cancers and leukemias.

The term "diagnostic method" as used herein means a method that enables a determination of the presence or type of cancer in or on a human or animal. Suitably the marker allows success of roscovitine treatment to be assessed. As discussed above, suitable diagnostics include probes directed to any of the genes as identified herein such as, for example, QPCR primers, FISH probes and so forth.

The present invention will now be described with reference to the following examples.

EXAMPLES:

MATERIALS AND METHODS Trial design and patient samples

Between 2001 and 2004, 56 patients with metastatic or locally advanced malignant solid tumours were recruited into a single agent seliciclib Phase Ib study. The patients had a range of tumour types, and were treated with seliciclib at different doses and schedules. During the first cycle of treatment, blood samples were taken on the first and last day of dosing, which represented 3, 5 or 10 days of consecutive dosing depending on the patient's treatment regime. Samples were collected in tubes containing sodium heparin as an anticoagulant; plasma was isolated by centrifugation, and aliquoted into fresh tubes prior to storage at -80°C. Of the 56 patients, 48 had suitable plasma samples taken before and after treatment. Informed consent for analysis of pharmacodynamic biomarkers was obtained from all patients.

Sample preparation

Plasma samples were diluted 1 :6 in U9 buffer (9 M urea, 2% CHAPS, 50 mM Tris- HCl pH 9), mixed for 30 minutes at room temperature and then further diluted 1:10 with the appropriate binding buffer. The surface of all ProteiaChip® Arrays (Ciphergen Biosystems, Fremont, CA) were pre-equilibrated with the appropriate binding buffer for 20 min, prior to the addition of samples: QLO buffer (100 mM Tris- HCl pH 9, 0.1% Triton X-100); CMlO buffer (100 mM NaOAc pH 3.5, 0.1% Triton X-100); H50 buffer (10% acetonitrile, 0.1% trifluoroacetic acid); the metal affinity IMAC chips were activated with 0.1 M cupric sulphate, then equilibrated in IMAC buffer (0.1 M sodium phosphate- pH 7, 0.5 M NaCl). Plasma samples were allowed to bind to the chip surfaces for 1 hour at room temperature on a platform shaker. Arrays were washed once with binding buffer, followed by two washes with binding buffer in the absence of detergent; all washes were performed for 5 minutes on a shaking platform. The chip surfaces were then rinsed briefly with 10 mM Hepes pH 7 and left to air-dry. A 50% saturated sinapinic acid matrix solution (prepared in 50% acetonitrile, 0.05% TFA) was appliedtwice (0.8 μl each time) to the chip surface to facilitate desorption and ionisation. Proteins retained on the different chip surfaces were ionised and detected in a ProteinChip® Reader (Ciphergen- Biosystems). Data, was collected using the ProteinChip® Software 3.1 (Ciphergen Biosystems) in three different mass ranges; 0-50,000 (low), 0-100,000 (mid) and 0-200,000 (high), by averaging 80-100 laser shots with an intensity of between 190 and 210.

Data analysis

The intensity of the detected peaks was normalized to the total ion current for each plasma sample. The Biomarker Wizard Software (Ciphergen Biosystems) clusters peaks of similar molecular weight from all the spectra, then displays changes in peak profiles between the different sample groups (untreated and treated plasma). AU spectra were compiled, and qualified mass peaks (signal-to-noise ratio >5) with mass- to-charge ratios (m/z) between 2000 and 200,000 were autodetected. Data collection for peak clusters was completed using second-pass peak selection (signal-to-noise ratio >2, within 0.3% mass window), and estimated peaks were added. The mean and standard deviation for each sample group was reported and the- Mann- Whitney U test was used to analyse the data and a p-value assigned to each cluster group that was deemed to be significantly different between the untreated and treated plasma samples.

Two-dimensional gel electrophoresis

For isoelectric focusing, plasma samples (5 μl) were initially loaded into pH 4-7 Immobiline DryStrips (Amersham Biosciences, Bucks, UK) by in-gel rehydration at room temperature for 16 hours using rehydration buffer (8 M urea, 2% CHAPS, bromophenol blue, 20 mM DTT and 0.5% IPG pH 4-7 buffer).- Proteins were then resolved in the IPGphor apparatus using a total of 40,000V over a period of 8 hours with a constant current of 50 mA per strip. After isoelectric focusing, the strips were equilibrated at room temperature for 30 minutes with equilibration buffer (50 mM Tris-HCl, pH 8.8, 6 M urea, 30% glycerol, 2% SDS, 0.01% Bromophenol blue and 10 mg/ml DTT), followed by a further 15 minutes in equilibration buffer with no DTT, but containing 25 mg/ml iodoacetamide. Second-dimensional size separation was performed using 4-12% polyacrylamide gradient gels (Iπvitrogen, Paisley, UK). Colloidal blue solution (Invitrogen) was used to stain proteins within the gel following electrophoresis. Protein spots of interest were excised from gels and processed in parallel for protein identification by trypsin digestion and to confirm their molecular weight by SELDI-TDF MS.

Protein identification

Gel pieces were incubated in 100 mM ammonium bicarbonate, 50% acetonitrile for three washes (10 minutes each) and then with 100% acetonitrile for 5 minutes. The gel pieces were removed and dehydrated by incubation for 5 minutes at 65⁰C. To digest the protein within each gel piece, porcine trypsin (Promega, Southampton, UK) was added (10 ng/μl in 25 mM ammonium bicarbonate) and incubated overnight at 37°C. The peptide digests were analysed by MS and MS/MS on an ABI 4700 Proteomics Analyzer with TOF/TOF Optics (Applied Biosystems, Warrington, UK). To identify each protein, the MS and MS/MS data from the peptide mass fingerprints of the digested gel spots were used to search sequence databases using the Mascot search engines from Matrix Science (http://www.matrixscience.com).

Elution of proteins from 2D gels

Gel pieces were incubated with 100% acetonitrile for 5 minutes on a shaker, the solution was removed and the gel pieces left. to dehydrate for 5 minutes at 65°C. FAPH solution (50% formic acid, 25% acetonitrile, 15% isopropanol, 10% water) was added to each gel piece and placed in a sonicating waterbath for 30 minutes at room temperature. Samples were vortexed for a further 2-3 hours, prior to the jeluate being applied to an NP20 chip (1-5 μl) and analysed in the ProteinChip® reader to confirm that the profile of the protein extracted from the gel piece aligned with the biomarker peak in the original SELDI-TOE MS spectra.

Immunocapture of proteins from plasma

RSlOO chips (Ciphergen Biosystems) have preactivated surfaces that contain carbonyl di-imidazole moieties, which react covalently with primary amine groups. Using this feature, antibodies can be bound to these surfaces thereby generating antibody chips that allow the capture oflspecific proteins from complex mixtures such as plasma. The- arrays were used as described in the manufacturer's instructions with the following antibodies; Transthyretin (DAKO, Cambridgeshire, UK)₅ Apolipoprotein AI and Apolipoprotein CI (both Chemicon, Hampshire, UK);. Briefly, 2 μl antibody (0.1 mg/ml) was applied- to the chip surface and any remaining active sites were blocked with BSA. Plasma samples were then applied irrPBS,-pH 7.2 containing 0.1 % Triton X-1Q0, and incubated for.2 hours at room temperature. Chips were rinsed with PBS, and urea CHAPS buffer (50 mM Tris pH 7.2, 1 M urea, 0.1% CHAPS₅ 0.5 M NaCl), and finally with Hepes pH 7. Chips were processed for analysis as described above. Treatment of mice

Immediately prior to dosing, compounds were prepared as a suspension in 50 mM HCl. Groups of five mice were orally administered 400 mg/kg b.i.d. seliciclib, 400 mg/kg b.i.d. D7-seliciclib or vehicle, for four consecutive days. No adverse effects were noted in any of the mice over the four days of treatment. Two hours after the final dose, mice were sacrificed and a terminal bleed was performed. The blood was heparinised, centrifuged and the plasma aspirated and frozen at -70°C.

RESULTS Proteomic profiling of plasma samples

SELDI-TOF MS combines protein purification chromatography with mass spectrometry to enable the analysis of proteomic patterns in complex biological samples. ProteinChip® Arrays consist of chemically treated surfaces that capture subsets of proteins from within a complex sample based on the biochemical properties of the individual proteins.

As part of a Phase Ib clinical trial, patients received doses of seliciclib ranging from 200 mg/day to 3600 mg/day, which were administered over 3, 5 or 10 consecutive days. Plasma samples, pre and post treatment, were collected during the first cycle of treatment from 48 of the 56 patients enrolled in the study. Plasma samples were profiled on four chip surfaces; CMlO, QlO, H50 and IMAC-Cu2+, which represent weak cation exchange, strong anion exchange, hydrophobic and metal affinity chip surfaces* respectively. Each sample was run in duplicate and the Biomarker Wizard software was then applied to the complete datarset for each^" chip surface to detect peaks that were modulated by seliciclib (192 spectra/surface). The individual spectra for all patients were then examined to verify .the- differences in plasma proteomic profiles highlighted by the Biomarker Wizard. A number of highly significant peaks were identified, that were most prominent on the _CM10 and QlO chip surfaces. The main peaks that appeared following treatment were located at approximately 7 kDa, 14 kDa, 17 kDa and 28 kDa (Fig. IA). Following seliciclib treatment two markers appeared on the CMlO chip surface in the ~7 kDa mass region. These two additional peaks at ~6.8 kDa and ~7 kDa occurred in the majority of the patients (Fig. IB), however in the three patients that received the lowest doses of seliciclib, the two peaks were weak (patient 106, 400 mg/day) or absent (patients 101 and 102, 200 mg/day). This data would suggest that the appearance of the peaks was related to the dose of the compound and required the administration of at least 400 mg/day. Figure 1C shows the Biomarker Wizard plot summarising the data in the 7 kDa mass region for all 48 patients along with representative spectra for two patients. The intensity of the two peaks at 6441 Da and 6639 Da appeared to decrease slightly following seliciclib treatment (P=O-Ol and p=0.5 respectively), whereas the peaks-at 6790 Da and 6990 Da increased significantly-

(both p<lE-10). Detailed analysis of the other three -markers identified by the

Biomarker Wizard indicated that in each region an equivalent reciprocal relationship was also evident for those biomarkers (data not shown). That is, whenever the software identified an increase in a specific peak, this was associated with the decrease in a peak of slightly smaller molecular weight. These results were perhaps indicative of a post-translational change occurring to each of the markers that was induced by seliciclib treatment. The first step to address this possibility was to identify the proteins that corresponded to these spectral peaks. Identification of biomarkers

Representative patient plasma was analysed by two-dimensional gel electrophoresis using pH 4-7 gradient strips for the first dimension and 4-12% polyacrylamide gels for- the second dimension. Examination of the stained gels highlighted two clusters^" of protein spots that differed between the pre and post treatment plasma samples, and which ran at molecular weights compatible with the 14 kDa anά-28 kDa biomarkers identified by the SELDI-TOF MS analysis (Fig._ 2A). An expanded view of the ~14 kDa protein spots that were modulated inrresponse to seliciclib treatment is shown (Fig. 2B). In both- these, patients an additional spot was observed in the post treatment sample, as indicatedLby the arrows. Gel pieces containing the -14 kDa protein spots of interest (labelled 1 and 2 in Fig. 2B) were excised from the gel, divided in half and processed in parallel in order to either elute the protein from the gel piece or to perform in-gel tryptic digestion. The eluted samples were applied to an NP20 chip (normal phase) and reanalysed on the ProteinChip® Reader to establish if they corresponded to the original biomarker peaks at ~14 kDa. Protein extracted from spot 1, which was present pre and post treatment, aligned with the first peak (→), which was also present pre and post treatment (Fig. 2C). Protein eluted from spot 2 on the 2006/004561

28

other hand, which was significantly enhanced following treatment, aligned with the novel ~14 kDa peak that appeared in the spectrum after 10 consecutive days of seliciclib treatment ("^). No protein was successfully eluted from the spot 2 region of the pre-treatment samples (data not shown). Peptide fragments released by tryptic digestion of the same gel piece were analysed by MS and MS/MS to reveal the peptide mass fingerprints of each sample whieh were then used for protein identification (data not shown). The protein eluted from both spot 1 and 2 in the 14 kDa region^"of the gel was identified as transthyretin. Since both spots were identified as transthyretin this suggested again that rather than the appearance of a new protein the post treatment markers were due to a post-translational modification of existing proteins, resulting in a more acidic form of that protein.

As final confirmation that, transthyretin was being modified during seliciclib treatment immunocapture experiments were performed. A transthyretin antibody was applied to an RSlOO preactivated chip surface, which was then used to specifically capture transthyretin from patient plasma samples. The transthyretin antibody specifically retained four peaks of ~14 kDa from plasma following treatment (Fig 3A bottom spectra). The^~first three peaks had molecular masses of 13,710, 13,846 and 14,048 Da and corresponded to transthyretin in its native, cysteinylated and glutathionylated forms respectively [19]. The fourth peak at 14,1-91 was novel to post seliciclib treatment and aligned with the novel marker peak that was identified- in the original- patient plasma profiling on the QlO chip (first spectra, 14,199). Since the mass accuracy of the PBSH^~reader is ~0.1% for external calibration, the 8^" Da difference observed between the two marker peaks on the different chip surfaces is within the mass error. These data thereby confirm that the ~14 kDa protein was transthyretin. It should be noted that in the original plasma profiling the additional peak in the patient spectra (*) is completely absent in the immunocapture spectra. This peak was seen in the profiling of all the patients (see also patient 209 Fig. 2C). It has been determined that this peak is a half mass peak of a larger protein, which explains why it is not retained in the transthyretin immunocapture experiment. As noted earlier there was a second cluster of protein spots that appeared to differ between the pre and post treatment gels and was in the general mass range of the 28 kDa marker identified by the SELDI-TOF MS analysis (Fig. IA, upper box Fig. 2 A

& Fig. 2D). In this region of the gel there were two distinct protein spots prior to seliciclib treatment that became three spots post treatment due to the appearance of a more acidic protein spot (arrow in Fig 2D). To identify this spot, similar approaches were taken to those described for transthyretin. These experiments indicated that the

28 kDa cluster of protein spots was Apolipoprotein AI (data.not shown).

No protein spots were observed on the polyacrylamide gels that could be assigned to either the 7 kDa or the 17 kDa markers identified by the SELDI-TOF MS analysis. Since transthyretin and apolipoprotein AI are both abundant plasma proteins, a literature search was performed to highlight additional plasma proteins of ~7 kDa and -17 kDa that could be the SELDI-TOF MS markers. One such candidate protein for the ~7 kDa peaks was Apolipoprotein CI, which occurs in two forms in human plasma; a 6.6 kDa form, corresponding to the full-length ApoCI protein, and a 6.4 kDa form, referred to as ApoCF which is truncated by two amino acids at the amino-terminus. The size of Apolipoprotein CI therefore appeared to be equivalent to the two peaks that decreased slightly in association with the appearance of the marker peaks at ~7 kDa (Fig. 1C).- In order to investigate this further, immunocapture experiments were performed to determine whether the ~7 kDa biomarker peaks were ApoCI (Fig. 3B). Specific peaks were observed when -purified ApoGI protein was applied to RSlOO antibody chips (Fig. 3B -second spectra) and these aligned with the peaks retained on? CMlO chips from patient plasma, indicating that ApoCI was a good candidate for this marker. Moreover, when patient plasma was applied-to the antibody chip, an additional peak (6994 Da) was retained on the post-treatment plasma chips that ialigned_with one of the marker peaks identified during the original profiling experiment. The biomarker peak at 6792 was always less intense than the larger marker and in this analysis appeared to be masked by a set of low intensity peaks when ApoCI was immunocaptured. Thus the ~7 kDa markers appeared to be the result of posttranslational modifications of ApoCI and ApoCI'.

In one particular experiment examining protein-protein interactions, samples were treated with DTT to disrupt disulphide bridges. Surprisingly, the cluster of peaks at

~17 kDa disappeared and a set of peaks at half that mass appeared (data not shown). By searching the literature, it became apparent that the most likely candidate for this set of peaks was Apolipoprotein All, which is found in human plasma as a disulphide-linked homodimer of 17.4 kDa.

Examining the nature of the modifications

Statistical analysis of the changes in the markers indicated that within patients there was a very strong positive relationship for all the markers as determined by their Pearson correlation coefficient. That is, within each patient there tended to be an equivalent increase in the intensity of each of the markers. Further analysis. of the mass changes associated with each marker revealed mat, in each case, the average increase in molecular weight of the novel peaks was -351 Da larger than the pre-existing peaks reduced in abundance following treatment. The equivalent size changes and the very strong correlation between the markers suggested that a similar modification was likely to be occurring on each of the proteins identified.

Since seliciclib has a molecular weight of 354 Da, one possible explanation for the mass shift of ~351 Da was that seliciclib or one of its metabolites was forming an adduct with the four plasma proteins. In order to investigate this hypothesis in more detail, plasma samples from a previous seliciclib xenograft experiment were analysed to determine if the same biomarkers were found in mice. Murine Apolipoprotein CI is five amino acid residues larger than its human equivalent, which was reflected in the spectra obtained from the ProteinChip® reader with the primary unmodified peak in the mouse plasma being slightly larger than theJiuman form (Fig 4A). As with the human plasma an additional peak was observed in-mice treated with seliciclib (Fig 4 A, second spectra) indicating that the same modification, leading to a ~351 Da increase in molecular weight, was occurring in treated mouse. To determine whether seliciclib was forming an adduct with the marker proteins, a D7 derivative of seliciclib was utilized that contains seven hydrogen atoms substituted by deuterium atoms, thereby making this compound 7 Da heavier than the parent compound (-361 Da). Three groups of non-tumour-bearing mice were treated for four consecutive days with 400 mg/kg seliciclib, D7-seliciclib or vehicle, administered twice daily. The spectra obtained from a representative mouse in each group are shown in Figure 4B. The comparison of the spectra from mice administered seliciclib and D7-seliciclib indicated that the mass shift was approximately 351 Da for seliciclib and 358 Da for D7-seliciclib. These data therefore suggest that the markers were due to the binding of seliciclib or one of its metabolites to the plasma proteins.

DISCUSSION

SELDI-TOF-MS analysis of the plasma samples highlighted several mass peaks that appeared in the plasma of the majority of patients following seliciclib treatment. Using a range of techniques, the proteins associated with the post treatment changes in SELDI-TOF spectra were subsequently identified as transthyretin, ApoAI, ApoCI and probably ApoAII, which are abundant plasma proteins. The kinetics of the appearance of the peaks was examined in more detail by profiling samples taken throughout the first day of treatment in three patients, who each received two doses of seliciclib per day. The biomarkers began to appear only after the second dose of seliciclib and in addition, the intensity of the peaks on day 1 was consistently less than was observed following 5 consecutive days of treatment (data not shown). The slow kinetics and repeated dose requirements suggested that the markers arose from a cumulative effect following treatment.

Pearson correlation coefficient analysis of the various plasma markers suggested that there was a very strong correlation between the different markers within the patients. Pearson values as high as 0.88 (p<0.0001) were obtained indicating that all the markers were potentially interrelated. This hypothesis was supported by the xealisation that each marker appeared to represent the gain of -351 Da by an abundant plasma- protein. Since seliciclib has a molecular weight of 354 Da it was possible that the novel markers appearing after treatment were a direct result of compound binding. Adduct formation was confirmed with the use of a deuterium substituted analogue of seliciclib (D7) that had a molecular weight of 361 Da and resulted in the appearance in D7 treated mice, of markers separated by 358 Da rather than the 351 Da seen in seliciclib treated mice.

Seliciclib is not inherently reactive, therefore the most likely candidates responsible for adduct formation are metabolites of seliciclib. The metabolism of seliciclib has been studied extensively, the principle phase I metabolite (PMF30-128) is formed through the sequential oxidation of the primary alcohol to the corresponding carboxylic acid via an aldehyde intermediate (Fig. 5). Several hypotheses can be envisaged to explain the appearance of the biomarker peaks. One possibility involves the direct coupling of the aldehyde, an intrinsically reactive species, to the proteins.

Preliminary experiments in vitro support the view that the aldehyde can become associated with proteins, however, the mass shift obtained in these experiments was only 345Da (data not shown):, which was smaller than that observed in patients. A second possibility is that the carboxylic acid metabolite, PMF30-128 interacts with plasma proteins either through direct esterification or via the formation of a reactive acyl glucuronide derivative during secondary metabolism. Since the markers appeared to be associated with compound binding the relationship between the markers and various pharmacokinetic parameters was examined. The most significant Pearson correlation coefficient values (e.g. 0.59 pO.OOOl) were found between the SELDI peaks and the Cmax/AUC for the carboxylate PMF30-128 on the last day of treatment. This was more significant than the correlation with plasma levels of the parent compound on the last day of treatment or indeed the levels of either PMF or seliciclib on the first day of treatment. This is in accordance with the earlier conclusion that repeated dosing and accumulation of the compound was necessary for the appearance of these peaks, and suggests PMF30-128 is the metabolite responsible for the adduct formation.

Transthyretin, apolipoprotein AI, CI and All are all abundant plasma, proteins. Importantly the presence of the compound-adducts did not appear, to be detrimental to the function of these proteins. Samples from several of the patients with significant marker peaks were analysed before and after treatment with regards to various clinical parameters to assess their function. There was no significant impact on any of these parameters as a result of the formation of compound adducts (data not shown). Moreover, no correlation was observed between the presence or abundance of the markers and any of the clinical parameters of toxicity measured in the patients, including elevations in creatinine and various liver enzymes (data not shown). The phenomenon of pharmaceutical agents binding to proteins has been widely recognised for some time. The best-studied example of a drug associated with adduct formation is Paracetamol (Acetaminophen). At therapeutic levels, paracetamol is mainly detoxified by glucuronidation and sulfation and excreted in the urine; a small amount, however, undergoes oxidation to form a reactive metabolite, N-acetyl-p- benzoquinoneimine (NAPQI), which becomes bound to -glutathione and is eliminated in the urine as glutathione conjugates. At higher doses, such as those seen with paracetamol overdose, the resultant increase in the production of NAPQI saturates the glutathione pools and the compound becomes bound to liver proteins. (Zhou, S., et al., Drug Metabolism Reviews, 2005. 1 : p. 41-213). Other drugs known to covalently bind proteins include antibacterial agents, such as sulphonamides, the anti-cancer tubulin- interacting drugs, as well as some steroids, NSAIDs and immunosuppressants such as cyclosporine A (reviewed in Zhou, S., et al., Drug Metabolism Reviews, 2005. 1: p. 41-213). Cyclosporine A is extensively metabolized by CYP3A4/3A5 to one or more reactive metabolites that can bind covalently to microsomal proteins. In the case of the NSAID' s, it is the reactive acyl glucuronide metabolite that is responsible for binding to albumin or other plasma proteins, as well as some cellular proteins in the liver or intestine. In contrast, for some anti-cancer drugs such as- paclitaxel and- the vinca alkaloids, it appears that it is the parent molecule rather than the metabolites' that are responsible for binding directly -to proteins. For example, many tubulin interacting drugs bind covalently to β-tubulin at residue Cys-239 in tumour cells, and this is -essential for their anti-cancer activity. With the advance in- technology provided by SELDI-TOF MS it will berpossible to analyse.plasma proteomic profiles following treatment with any drug to determine if adduct formation is^" observed. REFERENCES

1. McClue, S. J., et al., In vitro and in vivo antitumor properties of the cyclin dependent kinase inhibitor CYC202 (R-roscovitine). International Journal of Cancer, 2002. 102: p. 463-468. 2. Fischer, P .M-. and A. Gianella-Borr-aderi, CDK inhibitors in clinical development for the treatment of cancer. Expert Opin Iπvestig Drugs, 2003. 12(6): p. 955-70-.

3. Zarkowska, T. and S. Mittnacht, Differential phosphorylation of the retinoblastoma protein by Gl/S cyclin-dependent kinases. The Journal of Biological Chemistry, 1997. 272(19): p. 12738-46.

4. Xu, M., et al., Cyclin A/CDK2 binds directly to E2F-1 and inhibits the DNA- binding activity of E2F-1/DP-1 by phosphorylation. Molecular and Cellular Biology, 1994. 14(12): p. 8420-31.

5. Lolli, G., et aL, The Crystal Structure of Human CDKl and Its Protein Recognition Properties. Structure, 2004. 12(11): p. 2067-79.

6. Ramanathan, Y., et al., Three RNA polymerase TI carboxyl-terminal domain kinases display distinct substrate preferences: Journal.ofBϊological Chemistry, 2001. 276(14): p. 10913-10920.

7. Lam, L.T., et al., Genomic-sβale measurement of mRNA turnover and the mechanisms of action of the anticancer drug flavopiridoL

Genomebiology.Com, 2001. 2(10): p. 1465-6914.

8. Ljungman, M. and M.T. Paulsen, The cyclin-dependent kinase inhibitor roscovitine inhibits RNA synthesis and triggers nuclear accumulation ofp53 that is unmodified at Ser 15 andLys382. MoI Pharmacol, 2001. 60(4): p. 785-9. 9. MacCallum, D.E., et al., Seliciclib (CYC202, R-roscovitine) induces cell death in multiple myeloma cells by inhibition of RNA polymerase II dependent transcription and down regulation of McI-I. Cancer Research, 2005. 65(14): p. 5399-5407. 10. Negm, R.S., M. Verma, and S. Srivastava, The promise of biomarkers in cancer screening and detection. Trends in Molecular Medicine, 2002. 8(6): p. 288-293.

11. Sikora, K., Surrogate endpoints in cancer drug development. Drug Discovery Today, 2002. 7(18): p. 951-956.

1^"2. Petricoin, E.F., III, et al., Use of proteomic patterns in serum to identify ovarian cancer. The Lancet, 2002. 359: p. 572-577.

13. Rai, A. J., et al., Proteomic Approaches to Tumor Marker Discovery. Arch Pathol Lab Med, 2002. 126: p. 1518-1526. 14. Qu, Y., et al., Boosted Decision Tree Analysis of Surface-enhanced Laser Desorption/Ionization Mass Spectral Serum Profiles Discriminates Prostate Cancer from Noncancer Patients. Clin Chem, 2002. 48(10): p. 1835-1843.

15. Petricoin, E.F., III, et al., Serum Proteomic Patterns for Detection of Prostate Cancer. J Natl Cancer Inst, 2002. 94(20): p. 1576-1578. 16. Zhang, Z., et al., Three Biomarkers Identified from Serum Proteomic Analysis for the Detection of Early Stage Ovarian Cancer. Cancer Research, 2004. 64(16): p. 5882-5890.

17. Xiao, Z., et al., Serum. Proteomic Profiles Suggest Celecoxib-Modulated Targets and Response Predictors. Cancer Res, 2004. 64(8): p. 2904-2909. 18. Pusztai, L., et al., Pharmacoproteomic analysis of prechemotherapy and postchemotherapy plasma samples from patients receiving neoadjuvant or adjuvant chemotherapy for breast carcinoma. Cancer, 2004. 100: p. 1814- 1822.

19. Schweigert, F., K. Wirth, and J. Raila, Characterization of the microheterogeneity of transthyretin in plasma and urine using SELDI-TOF-MS immunoassay. Proteome Science, 2004. 2(1): p. 5.

20. Zhou, S., et al., Drug bioactivation, covalent binding to target proteins and toxicity relevance. Drug Metabolism Reviews, 2005. 1: p. 41-213. 21. Biven, K., et al., A novel assay for discovery and characterization of pro- apoptotic drugs and for monitoring apoptosis in patient sera. Apoptosis, 2003. 8(3): p. 263-8.

22. Ueno, T., et al., Measurement of an apoptotic product in the sera of breast cancer patients. European Journal of Cancer, 2003. 39(6): p. 769-774.

23. Kramer, G., et al., Differentiation between Cell Death Modes Using Measurements of Different Soluble Forms of Extracellular Cytokeratin 18. Cancer Res, 2004. 64(5): p. 1751-1756.

In preferred embodiments, the invention is preferably conducted in vitro. In these embodiments, preferably the methods do not involve intervention on the human or animal body, for example in the process of sample collection. In these embodiments, preferably the sample is provided as an in vitro sample.

All publications mentioned in the above specification, and references cited in said publications, are herein incorporated by reference. Various modifications and variations of the described methods and system of the present invention will be apparent to those skilled in the art without departing from the scope and spirit of the present invention. Although the invention has been described in connection with specific preferred embodiments, it should be understood that the invention as claimed should not be unduly limited to such specific embodiments. Indeed, various modifications of the described modes for carrying out the invention which are obvious to those skilled in molecular biology or related fields are intended to be within the scope of the following claims.

Claims

1. A method of monitoring activity of a CDKI comprising : a) isolating a sample, a "treated sample" from an animal model or human, wherein said animal model or human has been treated with a CDKI; b) determining the proteomic profile of the treated sample; c) comparing the -proteomic profile of the treated sample with a normal sample wherein an altered proteomic profile of the treated sample compared with the normal sample is an indication of CDKI activity.

2. A method as claimed in claim 1 wherein the sample is blood, serum or plasma.

3. A method as claimed in claim 2 wherein the proteomic profile is the plasma proteomic profile.

4. A method as claimed in claim 3 wherein the plasma proteomic profile is determined by SELDI-TOF MS or 2D PAGE.

5. A method as claimed in claim 3 or claim 4 wherein the-altered plasma proteomic profile is an alteration in mass peaks.

6. A method as. claimed in-any of claims 1 to 5 wherein the alteration in mass peaks is an alteration in an abundant plasma protein.

7. A method as claimed in claim 6 wherein the plasma protein is any of a 7, 14, 17 or a 28 kDa protein.

8. A method as claimed in claim 7 wherein the 7kDa protein is ApoCI, the 14kDa protein is transthyretin, the 17 kDa protein is ApoAII and the 28 kDa protein is

ApoAI.

9. A method as claimed in any of the preceding claims wherein alteration in mass peaks is a mass shift.

10. A method as claimed in any of the preceding claims wherein the alteration is an increase in molecular weight.

11. A method as claimed in any of the preceding claims wherein the alteration is the presence or absence of one or more post translational modifications of an abundant plasma protein in the treated sample compared to the normal sample.

12. A method as claimed in any of the preceding claims wherein the altered expression is due to the binding of a CDKI or a CDKI metabolite to an abundant plasma protein.

13. A method as claimed in claim 12 wherein the CDKI is seliciclib and the altered expression is due to the binding of a seliciclib metabolite.

14. A method as claimedrin claim 13 wherein the seliciclib metabolite is PMF30-128.

15. A method of monitoring activity of a CDKI comprising: a) isolating a sample, a "treated sample", from an animal model or human, wherein said animal-model or human has been-treated with a CDKI; b) determining altered expression of at least one of i) a 7 kDa protein or ii) a 17 kDa protein in said treated- sample as compared to an untreated control sample as an indication of CDKI activity.

16. A method as claimed in claim 15 wherein altered expression is an increase in the molecular weight of a 7 or 17 kDa protein.

17. A method as claimed in any of claims 15 or 16 wherein the 7 kDa protein is apolipoprotein Cl and the 17 kDa protein is apolipoprotein All.

18. A method as claimed in any of claims 15 to 17 wherein the sample is blood, plasma or serum.

19. A method as claimed in any preceding claim wherein the CDKI is seliciclib (roscovitine).

20. A method of assessing suitable dose levels of roscovitine comprising monitoring the altered expression of at least one abundant-plasma protein.

21. A method as claimed in claim 20 wherein the abundant plasma protein is a 7, 14, 17 or a 28 kDa protein.

22. A method as claimed in claim 21 wherein the 7kDa protein is ApoCI, the 14kDa protein is transthyretin, the 17 kDa protein is ApoAII and the 28 kDa protein is

ApoAI.

23. A method for identifying a candidate drug having CDKI-like activity comprising administering said candidate drug to an animal model or human and detecting altered. expression of at least one of a 7, 14, 17 or a 28 kDa protein in said treated sample as compared to a normal samplers an indication of CDKI activity.

24. A method- for identifying a candidate drug having_ CDKI-like activity comprising administering" said candidate drug to an animal model or human and detecting altered expression of at least one of i) a 7 kDa protein or ii) a 17 kDa protein in said treated sample as compared to a normal sample as an indication of CDKI activity.

25. Use of at least one of a 7 kDa or 17 kDa protein in the monitoring of activity of a CDKI.

26. A kit for assessing the activity of a CDKI comprising antibodies for at least one of a 7 kDa or 17 kDa protein,

27. A method as claimed in any of the preceding claims wherein roscovitine is administered to a human over a period of days prior to removing blood samples.

28. A method for identifying seliciclib (roscovitine) uptake in a patient comprising identifying plasma proteins associated withra-seliciclib metabolite.

29. A method for identifying seliciclib (roscovitine) uptake in a patient comprising identifying a shift in the mass of a plasma protein.