Particularly the present invention provides a method for discriminating dysplastic cells over- expressing INK4a gene products from cells over-expressing INK4a gene products without being dysplastic by detection of a marker suitable for characterising the proliferation properties of the respective cell. The characterisation of the proliferation properties may comprise the detection of a marker or a set of markers characteristic for active cell proliferation and/or a marker or a set of markers characteristic for retarded or ceased cell proliferation. The method presented herein thus enables for a specific diagnosis of dysplasias in histological and cytological specimens.

The detection of the over-expression of p1 6INK4a in biological samples has proven as a useful marker in the detection of anogenital lesions, such as carcinoma of the uterine cervix (see WO00/01845 ; Klaes et al., Int. J. Cancer: 92,276-284 (2001) ). The method based on p16lNK4tspecific immuno- chemical staining allows for a sensitive and specific identification of dysplastic cells in tissue section and in cytological samples.

In immuno-histochemical examinations of tissues dysplastic and neoplastic cells can be stained using a pl6lNKU specific antibody mediated staining procedure. The histological diagnosis of neoplastic lesions can thus be supported by a staining procedure based on a molecular marker characteristic for transformation of cells in anogenital lesions. The diagnosis, whether or not cells are neoplastic, in these procedures is not solely based on the p16lNK4a specific staining, but does also rely on the histological information.

This is due to the fact, that in about 20-30 % of samples metaplastic cells show some immuno- reactivity with p16§NK4a specific antibodies, and thus are stained in the course of the procedures. Yet the staining pattern yielded from these metaplastic cells differs from the pattern, which renders from neoplastic lesions. Metaplastic cells give rise to a patchy or focal staining pattern, whereas neoplastic lesions give rise to diffuse staining pattern. Moreover the staining-intensities of metaplastic cells are predominantly less than that of neoplastic cells.

The common methods used in screening tests for the early detection of dysplasias and/or neoplasias do not employ histology based tests, but do rather rely on cytological testing procedures. Yet especially in cases, when there is no histological information available concerning the architecture of tissues, such as for example in cytological examinations, testing for p16lNK4a over-expression alone

may lead to false positive results. This is due to the fact, that those fractions of metaplastic cells expressing p16lNK4a at detectably elevated levels, may not be differentiated by means of a histologic criteria.

The percentage of cells showing over-expression of p16lNK4a increases in the course of emergence of dysplasias. So in neoplastic or pre-neoplastic stages, when only a restricted population of neoplastic or pre-neoplastic cells is present in samples the immuno-reactivity of p16lNK4a may be weak. This weak immuno-reactivity may be of about the level as the level caused by metaplastic cells. In later stages of dysplasias the overall immuno-reactivity of p16lNK4a is stronger and so neoplastic lesions are easily discemible from metaplasias even in a cytological testing format. This might lead to cases, where the presence of metaplastic cells expressing p16lNK4a might be confused with the presence of neoplastic cells, and thus produces a false positive result.

Especially in screening tests, where the detection of early stages of neoplasias is desirable this condition is quite unpleasant. This is especially true, as the p16lNK4a based diagnosis has proven to be a valuable tool in histological examinations and the application in cytological based screening procedures would be able to enhance these established procedures.

To reduce false positive results in cytological testing formats and so to further enhance the fidelity of the p16lNK4a mediated diagnosis of anogenital lesions a method for discriminating the metaplasias from neoplastic and dysplastic lesions would be desirable. A method for the discrimination of metaplasias from neoplastic and pre-neoplastic lesions is provided within the embodiments claimed according to the present invention.

For supporting the discrimination of metaplasias from neoplastic lesions in testing procedures based on the over-expression of p16lNK4a a marker molecule would be desirable, that is expressed in neoplastic and/or pre-neoplastic cells and tissues and, which is not expressed simultaneously with INK4a gene products in one single metaplasUc cells.

A solution to the problem present in the art is provided by the methods claimed according to this invention. In the course of the experiments leading to the present invention the inventors have found, that a combination of detection of the presence or absence and/or the level of p16lNK4a in combination with at least one marker for cell proliferation such as e. g. Il67, Ki-S2, mcm5 or mcm2 may solve the problem present in the art.

In the art a couple of documents pertaining to the use of a combination of molecular markers for improved diagnosis of dysplasias is presented. In W00208764 a method for improved diagnosis of cervical malignancies is disclosed using a combination of an HPV marker and a marker for cell

proliferation or viral activity. p16lNK4a is mentioned in the context of this invention as a viral activity marker to be combined with HPV markers.

In EP1217377 a method for automated detection of cervical malignancies is disclosed mediated by detection of more than one marker molecules. Some defined marker combinations are named within the document. There is no disclosure in this document relating to the choice of suitable markers for a combination. The purpose of the combination in this application is improved fidelity of the automated analysis of staining pattems in biological cytological specimens. This document mentions combination of p16lNK4a with other tumor markers.

W002059616 discloses a method for detection of cell-cycle disturbances for improved diagnosis of cervical malignancies. The document discloses, that dysplastic cells exhibit disturbances in cell cycle control and may thus be identified by means of detection of cyclin E type proteins together with post G1 substances in cells.

Jeffrey Keatng in °Ki67, Cyclin E, and p16lNK4a Are Complimentary Surrogate Biomarkers for human papilloma Virus-Related Cervical Neoplasia" (American Joumal of Surgical Pathology 25 (7): 884- 891,2001) discloses the complementarity of the use of p16lNK4a, Ki67 and Cyclin E in the course of diagnosis of cervical dysplasias. The document refers to the problems of each single marker in the detection of dysplasias and states, that p16lNK4a in combination with Cyclin E may be suitable to overcome the drawbacks of the single marker molecules especially when cytological specimens are under examination. No disclosure teaching use of p16lNK4a in concert with a marker characteristic for proliferating cells such as Ki67 is given. The document does not teach the combination of p16lNK4a for use in diagnostic methods; moreover the disclosure pertains to a restricted use of Ki67 in cervical differential diagnosis and thus teaches away from the use of this marker in diagnosis of cervical malignancies.

In the art there is no disclosure teaching the use of a combination of p16lNK48 with a marker characteristic for cell proliferation for use in a diagnostic method for improved discrimination of p16lNK43 positive non-dysplastic cells from p16lNK4a positive dysplasias. W002059616 does not give a hint as to the use of p16lNK4a in the detection of dysplastic cell proliferation. EP1217377 does not teach a purpose for the combination of tumor markers in the course of the detection than the automation of the analysis process. There is no disclosure pertaining to the advantage of combinations for specific discrimination purposes such as discrimination of dysplastic cells from e. g. metaplasUc cells in cervical specimens.

The inventors of the present invention sought to overcome the drawback present in the art that p16lNK4a, over-expressed in various dysplasias, may also be detected in some other non-dysplastic cells. The discrimination between non-dysplastic p16lNK4a positive cells and dysplasbc cells over- expressing p16lNK4a may be based on the proliferation characteristics of the respective cells. In normally controlled cells p16lNK4a inhibits cdk4 and thus inhibits proliferation. In contrast in dysplastic cells this regulation is impaired. Thus p16lNK4a does despite its uncommonly high expression level not lead to an inhibition of the cell proliferation.

The inventors of the present invention found that dysplastic cells may be discriminated from cells exhibiting controlled cell proliferation by the simultaneous detection of pl6lNK4a with a marker characteristic for cell proliferation. Due to the fact, that in normal cells elevated levels of p16lNK4a inhibit cell proliferation, cells over-expressing p16lNK4a may be classified as being dysplastic provided they are exhibiting the characteristics of active cell proliferation.

The present invention relates to a method for discrimination of neoplastic, pre-neoplastic and/or dysplastic lesions from non-dysplastic cells showing elevated levels of p16lNK4a, in biological samples in a histological or cytological testing procedure based on the detection of the presence or absence of cells expressing p16lNK4a gene simultaneously with cell proliferation markers in said biological samples. Suitable markers for cell proliferation may be e. g. Ki67, Ki-S2, Ki-S5, mcm5 or mcm2. In one embodiment of the invention protein or mRNA of the markers for cell proliferation may serve as a marker for discrimination of metaplasias from eariy dysplastic or pre-neoplastic lesions in samples.

It is one aspect of the present invention to provide a method for discriminating dysplastic cells over- expressing INK4a gene products from other cells expressing INK4a gene products at a detectable level in biological samples comprising determining in a cytological or histological testing procedure the co-expression of at least two marker molecules in at least one single cell, wherein at least one marker molecule is an expression product encoded by the INK4a gene and at least one further marker molecule is a cell proliferation marker, wherein the over-expression of at least one INK4a gene product and expression of at least one marker for active cell proliferation at an immuno- chemically detectable level within said single cell is indicative of the dysplastic state of the cell and wherein the over-expression of at least one INK4a gene product and expression of at least one marker for senescence, terminal differentiation of cells, apoptosis or cell cycle arrest at a detectable level within said single cell is indicative of the non-dysplastic state of the cell.

A second aspect of the present invention relates to a test kit for determination of dysplasias in samples according to the method disclosed herein.

During the experiments leading to the present invention it was found, that under certain circumstances non-dysplastic cells may show immuno-reactivity for p16lNK4a. The inventors found that these cells exhibiting ordered control of cell proliferation in response to the elevated levels of the INK4a gene products are subject to growth arrest. Thus these individual cells do not show immuno- reactivity for markers of cell proliferation. In contrast transformed cells over-expressing p16lNK4a exhibit dysregulated control of cell proliferation and do not respond to the elevated level of pl6lNK4a by cessation of proliferation. Thus these dysplastic cell show simultaneous expression of p16lNK4a with markers for cell proliferation. The inventors thus found out, that a simultaneous detection of p16lNK4a with markers for cell proliferation may serve to discriminate dysplastic cells from arrested cells over-expressing p16lNK4a such as e. g. metaplastic cells.

The discovery that p16lNK4a over-expressed in various dysplasias may also be detected in some other non-dysplastic cells led the inventors of the presented method to establish a technique for discrimination between non-dysplastc p16lNK4a positive cells and dysplastic cells over-expressing p16lNK4a based on the proliferation characteristics of the respective cells. Whereas in normally controlled cells p16lNK4a inhibits cdk4 and thus inhibits cell proliferation, this is not true for dysplastic cells. Thus in dysplastic cells p16lNK4a does despite its uncommonly high expression level not lead to an inhibition of the cell proliferation.

The method disclosed herein is based on the fact that dysplastic cells may be discriminated from cells exhibiting normally controlled cell proliferation by the simultaneous detection of a INK4a gene product such as e. g. p16lNK4a with a marker characteristic for cell proliferation.

The term marker as well as marker molecules in general shall be used herein to pertain to proliferation marker gene expression products as well as to INK4a gene expression products.

The denominations given throughout this text for genes may in part relate to the genes or proteins as they have been discovered from any organism. In the context of the present invention this denomination shall confer to the respective homologue of the named markers in the organism which is particularly in question for a method as disclosed herein. In certain embodiments of the present invention this organism is a mammal and in one embodiment may be a human being. Such in one embodiment of the present invention the named markers shall be the human homologues of the respective denominated ones.

Generally throughout the text the term" (cell) proliferation marker"or"marker for cell proliferation"in the various grammatical forms is used to denominate proteins as well as nucleic acid markers. In case the protein name of a marker such as e. g."replication protein"is used herein, this use shall be

understood to be metonymically and pertain as well to the protein as to the nucleic acid marker molecules encoding the particular protein A marker useful according to the present invention may be any molecule transcribed from a gene or any molecule translated from such a transcript Thus gene product as used in the context of the present invention may comprise polynucleotides such as e. g. DNA or RNA and polypeptides such as proteins, proteoglycans, peptides etc. Expression product (s) as used in the context of the present invention shall comprise any transcript of a gene locus in forward or reverse direction including any reading frames, splicing variants. Expression products as used herein shall thus comprise any alternative products encoded by the nucleic acids of a particular gene locus.

INK4a encoded gene-products as used in the context of the present invention shall be any mRNA transcribed from the INK4a gene locus or any polypeptide translated from such an mRNA. In one embodiment of the invention the expression products encoded by the INK4a gene may exhibit molecular weights of about 5 to 40 kDa or any value in between and preferably of about 10 to 20 kDa or any value in between and most preferably of about 14 to about 19 kDa or any value in between respectively.

INK4a gene-products suitable for the method according to the present invention may comprise e. g. gene-products such as e. g. p16 and p14ARF.

The term" (cell) proliferation marker"or"marker for cell proliferation"as used in the context of the present invention shall comprise any marker molecule known in the art to be characteristic for the proliferation status of cells. The proliferation status may e. g. be a status of actively proliferating cells, of retarded cell proliferation, of arrested cell proliferation, of senescent cells, of terminally differentiated cells, of apoptosis etc. In one embodiment of the invention the cell proliferation marker is a marker molecule characteristic for active cell proliferation. In another embodiment of the invention the proliferation marker molecule may be a molecule characteristic for arrested, terminally differentiated, senescent or apoptotic cells.

In certain embodiments proliferation markers for use in the context of the present invention may comprise genes engaged in the DNA replication such as e. g. proteins of the pre-initiation complex or of the replication fork. Such molecules may e. g. comprise helicases, such as eucaryotic helicase or MCM proteins (MCM2, MCM3, MCM4, MCM5, MCM6, MCM7), protein TP as disclosed in W00050451 and W00217947 (also denominated HELAD1, Pomfil2, Unc-53), kinases or phosphatases engaged in the replication process such as e. g. CDC6, CDC7 protein kinase, Dbf4, CDC14 protein phosphatase, CDC45 and MCM10. Furthermore proliferation markers may comprise

proteins engaged in the processive replication fork such as e. g. PCNA or DNA polymerase delta, replication protein A (RPA), replication factor C (RFC), FEN1.

In other embodiments the proliferation markers may comprise molecules necessary for the maintenance of cell proliferation such as Ki67. Ki-S5 or Ki-S2. In this embodiment proteins may be e. g. present throughout the whole cell cycle. They are useful for performing a method according to the present invention provided they are characteristic for active cell proliferation and are not significantly expressed in arrested, terminally differentiated, apoptotic or senescent states of cells.

Ki67, Ki-S2 and ki-S5 as used herein shall denominate the protein marker molecules detected by the respective antibodies as well as the nucleic acids encoding these antigens.

In another embodiment the cell proliferation markers for use in a method according to the present invention may be a marker molecule characteristic for retarded or ceased cell proliferation such as e. g. a senescence marker, a cell cycle arrest marker, a marker characteristic for terminally differentiated cells or an apoptosis marker. Such molecules comprise e. g. p21, p27, Caspases, BAD, CD95, fas-ligand, parp-proteins etc.

Discrimination as used in the context of the present invention shall comprise an assessment whether a sample is to be classified in one or another way. In one embodiment of the invention the discrimination pertains to the assessment of a tissue or components thereof being dysplastic or being not dysplastic. Thus the discrimination as used herein is a judgement about the growth properties of cells in a biological sample.

In one embodiment of the present invention the discrimination comprises the detection of expression of an INK4a gene product simultaneously with the detection of expression of a marker characteristic for active cell proliferation. In this case cells co-expressing both marker molecules are to be classified as being dysplastc.

In another embodiment of the present invention the discrimination comprises the detection of an INK4a gene product simultaneously with the detection of expression of a marker characteristic for arrested, ceased or retarded cell proliferation. In this case cells co-expressing both marker molecules are to be classified as being non-dysplastic.

In certain embodiments of the present invention it will be useful to detect the presence or absence and or the level of more than two marker molecules. In one embodiment one INK4a gene expression product will be detected in concert with two or more markers for cell proliferation. This may be useful to enhance the identification of proliferation properties of the INK4a gene product expressing cells in samples. Some proliferation markers are restricted to specific phases of the cell cycle or are present

in low abundance in cells. Due to this fact in some cases the detection of proliferating cells expressing INK4a gene products may be improved by detection of two or more proliferation markers.

In such cases e. g. one proliferation being expressed during the whole proliferative cell cycle may be detected simultaneously to markers being characteristic for specific cells cycle phases. E. g. Il67, Ki- S5 or Ki-S2 may be detected together with mcm5, mcm2, PCNA, rpA, rfC etc.. In other cases e. g. proteins engaged in the DNA replication may be detected together with Ki67, Ki-S5 or Ki-S2. In yet another case Ki67 may be detected together with Ki-S2. It must be understood, that these examples are intended to exemplify combinatorial possibilities and shall not be comprehensive, so that various other combinations of proliferation markers are likewise useful and suitable in the procedure of the method according to the present invention.

As the case may be, combination of two or more cell proliferation marker molecules may be applied for a method as disclosed herein. In another embodiment two or more marker molecules being detectable over larger stretches of the cell cycle or even over the whole cell cycle or actively proliferating cell may be detected in concert in a method as disclosed herein. A combination of more than one cell proliferation marker molecules may generally be useful for improving the sensitivity of the detection of the proliferation characteristics of cells.

In certain embodiments of the present invention a combination may furthermore comprise other marker molecules such as senescence marker molecules, markers for arrested cells, markers for terminally differentiated cells, markers for apoptotic cells, markers for viral infection or for viral activity in cells or cell cycle regulatory proteins as markers. In certain embodiments in connection with dysplasias being associated with HPV infection a detection of HPV associated marker molecules or markers for viral activity may be of use for a detection of a dysplasia. The methods useful for detection of detection of HPV infection in samples are known to those of skill in the art. These methods may comprise methods employing probes specific for HPV agents or may employ nucleic acid amplification reactions. The detection of a viral infection may be carried out simultaneously or subsequently to the detection of the INK4a and proliferation marker molecules.

Dysplastic as used in the context of the present invention shall refer to dysplasias from mild to severe dysplasias and their precursory stages, as well as carcinoma such as carcinomas in situ or invasive carcinomas and disseminated tumor cells. Thus dysplastic as used herein shall also comprise early and precursory stages of dysplasias and carcinomas.

Cells over-expressing INK4a gene products without being dysplastic (non-dysplastic as used herein) as mentioned herein may e. g. comprise metaplastic cells, senescent cells, terminally differentiated cells or cells, that in certain stages of the cell cycle exhibit elevated levels of the INK4a gene

products. In certain cells elevated levels of INK4a gene products may even be effected as a response to extemal signals such as hormones, transmitters etc. In one embodiment of the present invention the non-dysplastic cells over-expressing INK4a gene products comprise e. g. metaplasUc cells, endometrial cells etc.

The method for detection of the expression level of the INK4a encoded gene-products and/or the proliferation marker gene products according to the present invention is any method, which may (but need not) be e. g. suited to detect even very small amounts of specific biological molecules in biological samples. The detection reaction according to the present invention is a detection either on the level of nucleic acids or on the level of polypeptides.

A marker molecule is said to be detectable as used in the context of the present invention, provided the marker may be detected in the course of suitable detection procedure such as e. g. in-situ- hybridization, immuno-chemical staining, hybrid capture assay etc.. The level of expression of a marker molecule may be made detectable using suitable reporter reactions such as e. g. a chromogenic or fluorescence based immuno-chemical staining or in-situ-hybridizabon procedure for microscopic or automated analysis. Suitable methods for enhancing the reporter signal known to those of skill in the art may be applied in the course of a method according to the present invention.

Thus the marker is said to be detectable in a case where the staining supersedes the respective background staining inherently obtained in the immuno-chemical staining procedure so as to produce significant staining results.

The marker molecules may be detected using reagents that specifically recognise these molecules.

The detection reaction for the INK4a gene-products and/or the proliferation marker gene products may comprise one or more reactions with detecting agents either recognising the initial marker molecules or recognising the prior molecules used to recognise other molecules.

In certain embodiments of the present invention two or more probes may be used for the detection of one single marker molecule. For example two or more different binding agents (e. g. antibodies) or oligonucleotide probes directed against one single marker molecule (as the case may be against different epitopes or different sequences) may be used in the course of the method as disclosed herein.

The detection of the different gene products may be performed in one reaction vessel or containment or in different containments simultaneously or subsequently in time. Thus the different gene products may be detected simultaneously in one cell co-expressing both products. Otherwise cells co- expressing the gene products may be used for separated detection reaction (separated in space or

in time) to detect each a single marker in the cells. In another embodiment there might be cells expressing one or the other marker. The detection of the marker molecules in the different cells may as well be performed simultaneously or separately in time and/or space.

The detection reaction further may comprise a reporter reaction indicating the presence or absence and/or the level of the marker molecule gene-products. The reporter reaction may be for example a reaction producing a coloured compound, a bioluminescence reaction, a fluorescence reaction, generally a radiation emitting reaction etc..

In certain embodiments, different marker molecules may be recognised by agents, that produce different reporter signals, so that the signals referring to marker molecules could be distinguished. In one preferred embodiment of the invention the detection of the expression of the two or more INK4a gene-products and/or proliferation marker gene products is carried out simultaneously. In this case the reporter reaction may for example employ different fluorescent labels for the different molecules detected.

However within the context of the present invention it must not necessarily be answered whether the one or the other proliferation marker or INK4a marker gene product is expressed in the cells. In certain embodiments the question is whether any proliferation marker and/or INK4a gene product is expressed. Such in the course of the experiments a procedure may be chosen, that gives the same fluorescence signal as indication of the presence of a proliferation marker. This procedure is suitable to improve sensitivity of the detection of the cell proliferation characteristics (different markers characteristic for active cell proliferation). As the case may be the procedure may be applied as to render one detectable signal for three, four or even more marker molecules being characteristic for cell proliferation. Analogous the same may under certain circumstances be true for the INK4a gene expression products. It must be understood, that as the case may be different staining signals for different proliferation marker molecules may be desirable. The procedures may be applied to the necessities of the respective experiment.

In certain embodiments of the present invention a combination of one or more (e. g. two different) INK4a gene products may be detected with a combination of one or more e. g. a set of two, a set of three, a set of four, a set of five or a set of even more markers for cells proliferation. As the case may be the detection of the marker molecules for cell proliferation may render only one reporter signal. In other cases each single marker for cell proliferation may render a specific reporter signal or groups of marker molecules may render specific reporter signals.

Signals for the indication of the presence of immuno-reactivity may be chromogenic signals produced by various methods known in the art. Altematively or even in combination fluorescent signals may be used. Suitable reporter signals comprise fluorescent labels such as fluorescein, rhodamin etc.

Applicable formats for the detection reaction according to the present invention may be, blotting techniques, such as Westem-Blot, Southem-blot, Northem-blot, immuno-cytochemical or immuno- histochemical procedures. The blotting techniques are known to those of ordinary skill in the art and may be performed for example as electro-blots, semidry-blots, vacuum-blots or dot-blots. Immuno- cyto/histochemical staining procedures are known to those of skill in the art and may comprise binding agent mediated detection of polypepEdes as well as in situ hybridisation techniques. Both different techniques may even be applied simultaneously. In certain embodiment hybrid capture of nucleic acids may be used for the detection. Amplification reaction may also be applicable for the detection of e. g. nucleic acid molecules.

In one embodiment of the invention the detection of the level of INK4a and/or proliferation marker gene-products is carried out by detection of the respective nucleic acids (e. g. mRNA) or fragments thereof present in the sample. The means for detection of nucleic acid molecules are known to those skilled in the art. The procedure for the detection of nucleic acids can for example be carried out by a binding reaction of the molecule to be detected to complementary nucleic acid probes, proteins with binding specificity for the nucleic acids or any other entities specifically recognising and binding to said nucleic acids. In one embodiment in situ hybridisation of oligonucleotide probes to nucleic acids in a sample may be used for the detection of expression products or markers.

Probes as used in the context of the present invention may be any agent binding specifically to a molecule. In the case of nucleic acids a probe may be an oligonucleotide hybridising to a particular sequence. In one embodiment the probe may be e. g. a primer. In the case of the detection of polypeptides or proteins the probe as used herein may be e. g. a binding agent such as an antibody.

In certain embodiments of the present invention the probes may be detectably labelled. The label may be selected from the group comprising a radioisotope, a bioluminescent compound, a chemiluminescent compound, a fluorescent compound, a metal chelate, or an enzyme. Probes may be applied in any detection procedure known in the art e. g. in the course of an in situ hybridisation procedure, in the course of hybrid capture assays, in the course of immuno-chemical staining reaction, in the course of blotting techniques etc..

This method may be performed as well in vitro as directly in situ for example in the course of a detecting staining reaction. Another way of detecting the marker mRNAs in a sample performed in

the method according to the present invention is an amplification reaction of nucleic acids, which can be carried out in a quantitative manner such as for example the polymerase chain reaction. In a preferred embodiment of the present invention real time RT PCR may be used to quantify the level of marker mRNAs in samples of dysplasias or tumors (cells or tissue samples).

In another preferred embodiment of the invention the detection of the level of INK4a and/or proliferation marker gene-products is carried out by determining the level of expression of a protein or fragments thereof. The determination of the marker gene-product on the protein level can for example be carried out in a reaction comprising a binding agent specific for the detection of the particular marker polypeptide.

The binding agents can be used in many different detection techniques for example in westem-blot, ELISA or immuno-precipitation. Generally polypeptide binding agent based detection can be carried out as well in vitro as directly in situ for example in the course of an immuno-chemical staining reaction. Any other method for determining the amount of particular polypeptides in biological samples can be used according to the present invention.

The immuno-cytochemical imaging procedures for use in the context of the present invention may comprise e. g. the staining of cytological or histological preparations with chromogenic or fluorescent dyes. The staining may e. g. comprise binding of the molecules to be detected by a first binding agent, which itself is detected by a secondary binding agent, which may be labelled. The first binding agent may in certain embodiments be a nucleic acid or a protein binding agent (e. g. an antibody) and the secondary binding agent may be e. g. a secondary antibody recognizing the first binding agent.

Any methods known in the art for performing staining of cytochemical or histochemical staining may be applied in the course of a method according to the present invention.

Binding agents as used in the context of the present invention for the detection of the level of INK4a polypeptides such as p16lNK4a or p14ARF polypeptides and proliferation marker polypeptides such as e. g. mcm5, mcm2, Kl67, Ki-S5, PCNA or Ki-S2 polypeptides may comprise antibodies and antigen- binding fragments, bi-functional hybrid antibodies, peptidomimetics containing minimal antigen- binding epitopes, anti-cullines (anti-calineTM) etc.

An antibody or antigen-binding agent is said to react specifically, if it reacts at a detectable level with a protein disclosed herein, and does not significantly react with other proteins. The antibodies according to the present invention may be monoclonal or polyclonal antibodies. As used herein, the term antibody or monoclonal antibody is meant to include intact molecules as well as antibody

fragments. Moreover, antibodies of the present invention include chimeric, single chain, and humanised antibodies.

According to the present invention binding agents may be used isolated or in combination. By means of combination it is possible to achieve a higher degree of sensitivity. The term antibody, preferably, relates to antibodies which consist essentially of pooled monoclonal antibodies with different epitopic specificities, as well as distinct monoclonal antibody preparations.

Monoclonal antibodies are made from antigen containing fragments of the polypeptide of the invention using any of a variety of techniques known to those of ordinary skill in the art; see, e. g., Hariow and Lane, Antibodies: A Laboratory Manual, Cold Spring Harbor Laboratory, 1988. In one such technique, an immunogen comprising the antigenic polypeptide or a synthetic part thereof is initially injected into any of a wide variety of mammals (e. g., mice, rats, rabbits, sheep and goats). In this step, the polypeptides of this invention may serve as the immunogen without modification.

Altematively, particularly for relatively short polypeptides, a superior immune response may be elicited if the polypeptide is joined to a carrier protein, such as bovine serum albumin or keyhole limpet hemocyanin. The immunogen is injected into the animal host, preferably according to a predetermined schedule incorporating one or more booster immunizations, and the animals are bled periodically. Polyclonal antibodies specific for the polypeptide may then be purified from such antisera by, for example, affinity chromatography using the polypeptide coupled to a suitable solid support.

Within the context of the present invention it must not necessarily be answered whether the one or the other proliferation marker is expressed in the cells. In certain embodiments main question may be whether any proliferation marker is expressed. Such in the course of the experiments a procedure was chosen, that gives the same fluorescence signal as indication of the presence of a proliferation marker. This procedure is suitable to improve sensitivity of the detection of the cell proliferation characteristics. As the case may be the procedure may be applied as to render one detectable signal for three, four or even more marker molecules being characteristic for cell proliferation. Analogously, the same may under certain circumstances be true for the INK4a gene expression products. it must be understood, that, as the case may be, different staining signals for different proliferation marker molecules may be desirable. The procedures may be applied to the necessities of the respective experiment.

The INK4a gene-products and/or proliferation marker gene products may according to the present invention be detected simultaneously. In this context simultaneously according to the present

invention shall mean either literally at the same instant or within the same testing procedure, whereby the single detection steps are temporarily consecutive.

The detection procedure according to the present invention may furthermore comprise a cytochemical staining procedure rendering a chromogenic or fluorescent staining of cells or cell compartments. Such staining procedures are known to those of skill in the art and may for example comprise e. g. staining for acidophilic or basophilic structures, of subcellular regions (e. g. the nucleus, the mitochondria, the golgi, the cytoplasm etc. ), of specific molecules (the chormosomes, of lipids, of<BR> glycoproteins, of polysaccharids etc. ) in the cytological specimens. Fluorescence dyes such as DAPI, Quinacrin, Chromomycin, etc. may be employed. Furthermore chromogenic dyes such as Azan, Acridin-orange, Hematoxylin, Eosin, Sudan-red, Thiazin-stains (Toluidin-blue, Thionin) may be applied. In other embodiments staining procedures such as Pap-staining, Giemsa-staining, Hematoxylin-Eosin staining, van-Gieson staining, Schiff-staining (using Schiff reagent), staining procedures employing precipitation of metals (such as e. g. of silver in staining procedures employing Silver Nitrate) or insoluble stains such as e. g. of Tumbulls-blue (or other insoluble metal cyanides), etc. may be used in the course of a method as disclosed herein. It must be understood, that the named dyes and staining methods shall be examples for the applicable methods and that any other method known in the art may be applied to a method as disclosed herein.

The staining procedures may produce chromogenic stains for light microscopic inspection or fluorescent stains for inspection under fluorescence microscopic conditions. In another embodiment of the present invention radiation emitting procedures, procedures employing substances impairing the transmission of radiation or other contrast media for imaging of the cytological conditions in a sample (e. g. the generation of optical impression by means such as (micro-) autoradiographic or (micro-) radiographic picture generation) may be of use for a method according to the present invention.

All the staining and imaging procedures may be used for analysis not only in microscopic procedures but also in automated analysis procedures such flow cytometry, automated microscopic (computerized or computer aided) analysis or any other method for analysis of stained cytological specimens.

The analysis of the staining or imaging results of the different procedures may be performed in a single analysis step or in different subsequent steps. E. g. the light microscopic inspection of a specimen may be performed before or after fluorescence microsopic inspection of the specimen. In Fluorescence microscopy the analysis of different stains with different excitation wavelengths may be

analyses simultaneous or subsequently. Other imaging methods may be employed simultaneously or subsequently to the named procedures.

There may be various circumstances, under which combinations of different staining methods will be suitable. E. g. in cases, where no satisfying cytological staining results may be achieved by immuno- chemical staining the additional application of general cytological staining techniques may be suitable.

A sample according to the method of the present invention may comprise any sample comprising cells. Samples may comprise e. g. secretions, smears, body fluids, and cell-or tissue-samples.

In one embodiment of the present invention samples comprise cells of the anogenital tract, of the respiratory tract or of the skin and its appendages. In certain embodiments the cells may be cells of the uterine cervix, the vagina, the vulva, the penis, the anus, the rectum, the bronchic tree, the lung, the peritoneum, the peritoneal space, the naso-pharyngeal space, the oral cavity or the skin. In certain embodiments of the present invention the sample may be a histological sample, a biopsy, or a cytological sample such as e. g. a smear, a swab, a wash, a body fluid containing cells (sputum, a secretion, saliva, etc.). In certain embodiments of the present invention the samples may comprise cells infected by papilloma virus. The samples may in certain embodiments comprise cervical smears, bronchioalveolar lavages etc.

In certain special embodiments of the present invention the sample may be prepared as a monolayer or thin layer preparation of a cytological specimen. The respective methods for preparation of monolayer or thin-layer preparation in cytology are known to those of skill in the art. In one embodiment the preparation may e. g. comprise the ThinPrep technology. Other methods comprise conventional smears, or method employing suspensions of cells for preparation of the cytological specimens.

Preparation of a sample may comprise e. g. obtaining a sample of a tissue, of a body fluid, of cells from a patient. According to the present invention preparation of the sample may also comprise several steps of further preparations of the sample, such as preparation of dissections, preparation of cell suspensions, spreading or applying the cells to be examined onto microscopic slides, preparation of tissue arrays, isolation of polypeptides or nucleic acids, preparation of solid phase fixed peptides or nucleic acids or preparation of beads, membranes or slides to which the molecules to be determined are coupled covalently or non-covalenty.

In certain embodiments of the present invention the method may be performed in an automated manner. The automation of the method may be achieved by automated staining and analysis of

histological or cytological specimens on a solid surface by microscopic means. In another embodiment the automation may comprise a flow-cytometric analysis of the staining of cells in solution.

The dysplastic lesions to which the method according to the present invention may be applied comprise any dysplastic lesions characterized by over-expression of INK4a gene products such as e. g. p16lNK4a or p14ARF. In certain embodiments those lesions are dysplasias associated with infections by papilloma viruses such as e. g. HPV. In one embodiment the HPV may be a high risk HPV subtype such as HPV16, HPV18, HPV31, HPV 33, HPV35, HPV 39, HPV 45, HPV 51, HPV 52, HPV56, HPV 58, HPV 59, HPV 66, HPV 68 etc. In one embodiment the dysplastic lesions that may be detected according to the present invention comprise anogenital lesions, lesions of the respiratory tract, lesions of the head and the neck or lesions of the skin and its appendages. Such lesions may e. g. comprise dysplasias of the anus or rectum, of the vulva, the vagina, the cervix or the penis, of the bronchic tree, the lung, the oral cavity, or the nasopharyngeal space.

Another aspect of the present invention is a testing kit for performing the method according to the present invention. The kit may be for example a diagnostic kit or a research kit.

A kit according to the present invention comprises at least an agent suitable for detecting the INK4a gene-products.

Thus a kit according to present invention may comprise: reagents for the detection of one or more INK4a gene-products reagents for the detection of one or more proliferation marker gene-products reagents and buffers commonly used for carrying out the detection reaction, such as buffers, reporters reactants (dyes, etc.), carrier substances and others INK4a gene product samples for carrying out a positive control reaction Proliferation marker gene product samples for carrying out a positive control reaction.

The reagents for the detection of the marker gene-products may include any agent capable of binding to the marker gene-products. Such reagents may include proteins, polypeptides, nucleic acids, peptide nucleic acids, glycoproteins, proteoglycans, polysaccharids or lipids.

The INK4a gene-product and proliferation marker gene product samples for carrying out a positive control may comprise for example nucleic acids in applicable form, such as solution or salt, peptides in applicable form, tissue section samples or positive cells.

In a preferred embodiment of the invention the detection of the marker gene-products is carried out on the level of polypeptides. In this embodiment the binding agents may be for example antibodies specific for the marker gene-products or fragments thereof.

In another embodiment of the test kit the detection of the marker gene-products is carried out on the nucleic acid level. In this embodiment of the invention the reagent for the detection may be for example a nucleic acid probe or a primer reverse-complementary to said marker nucleic acids.

The present invention provides a method for the discrimination of neoplastic and/or dysplastic and pre-neoplastic lesions, identifiable by assessment of the over-expression of p16" fm other cells cells, which also detectably express p16lNK4a, in the course of histological and/or cytological testing procedures. The method is based on the detection of expressed gene-products of two or more INK4a gene products.

Thus the problem to be solved was to provide a method for discrimination between dysplastic cells and other cells lacking malignant growth potential. The method may be applied to any stage of dysplasias and may be especially useful in early stages, when cytological diagnostic methods based on the p16a over-expression needs a further information for the identification of metaplastic cells.

Furthermore the present invention provides a kit for performing the method according to the present invention.

Brief description of the drawings Figure 1: Fluorescent staining of a histological specimen of the cervix uteri; Figure 1 shows staining of a severe dysplasia using antibodies directed against p16lNK4a ; for experimental details see Example 1; immunoreactivity for p16lNK4a renders green fluorescence. Almost all cells of the lesion are diffusely positive stained in the cytoplasm and in the nuclei.

Figure 2: Fluorescent staining of a histological specimen of the cervix uteri; Figure 2 shows staining of a severe dysplasia using antibodies directed against Ki67; for experimental details see Example 1; immunoreactivity for Ki67 renders red fluorescence. Many cells of the dysplasia show nuclear positivity for Ki67.

Figure 3: Fluorescent double staining of a histological specimen of the cervix uteri; Figure 3 shows staining of a severe dysplasia using antibodies directed against Ki67 and those directed against p16lNK4a ; for experimental details see Example 1; immunoreactivity for Ki67 renders red fluorescence ; immunoreactivity for p16lNK4a

renders green fluorescence ; overlay of red and green fluorescence renders yellow fluorescence. Many cells of the dysplasia show nuclear positivity for Ki67 as well as positivity for p16lNK4a and give thus rise to yellow fluorescence.

Figure 4: Fluorescent staining of a histological specimen of the cervix uteri; Figure 4 shows staining of a squamous metaplasia using antibodies directed against p16lNK4a ; for experimental details see Example 1; immunoreactivity for p16lNK4a renders green fluorescence. Some cells of the metaplasia stain diffusely positive in the cytoplasm and in the nuclei.

Figure 5: Fluorescent staining of a histological specimen of the cervix uteri; Figure 5 shows staining of a squamous metaplasia using antibodies directed against Ki67 ; for experimental details see Example 1; immunoreactivity for Ki67 renders red fluorescence. Many cells of the squamous metaplasia show nuclear positivity for Ki67. Areas, that have been identified as being positively expressing p16lNK4a do not show any positive staining for Ki67 indicating lack of expression of the antigen in these areas.

Figure 6: Fluorescent double staining of a histological specimen of the cervix uteri; Figure 6 shows staining of a squamous metaplasia using antibodies directed against Ki67 and those directed against p16lNK4a ; for experimental details see Example 1; immunoreactivity for Ki67 renders red fluorescence ; immunoreactivity for p16lNK4a renders green fluorescence ; overlay of red and green fluorescence renders yellow fluorescence. Areas, that are positively expressing p16lNK4a do not show any positive staining for Ki67 indicating lack of expression of the antigen in these areas. No cells in the specimen give rise to yellow fluorescence.

Figure 7: Chromogenic double staining of a histological specimen of the cervix uteri; Figure 7 shows staining of a severe dysplasia using antibodies directed against Ki67 and those directed against p16lNK4a ; for experimental details see Example 6; immunoreactivity for Ki67 renders red nuclear staining; immunoreactivity for pl6lNK4a renders brownish staining over the whole cell ; double staining renders brown cells with red nuclei. Many cells of the dysplasia show nuclear positivity for Ki67 as well as positivity for p16lNK4a and give thus rise to a pattern of brown cells with red nuclei.

Figure 8: Chromogenic double staining of a cytological specimen of the cervix uteri; Figure 8 shows staining of a severe dysplasia using antibodies directed against Ki67 and those directed against pl6lNK4a ; for experimental details see Example 7; immunoreactivity for Ki67 renders red nuclear staining; immunoreactivity for p16lNK4a renders brownish staining over the whole cell ; double staining renders brown cells with red nuclei. Many cells of the dysplasia show nuclear positivity for Ki67 as well as positivity for p16lNK4a and give thus rise to a pattern of brown cells with red nuclei.

The following examples are given for the purpose of illustration only and are not intended to limit the scope of the invention disclosed herein. For the purpose of illustration the methods disclosed herein are exemplified using histological preparations. The histological examples aid to judge whether the cells stained in the one or the other way are to be classified as dysplastic or metaplastic. The methods may easily be transferred to cytological specimens by altering the protocol in appropriate manner. These alterations are known to those of ordinary skill in the art.

Example 1: Immuno-fluorescent detection of the over-expression of p16'Nx4a and Ki67in samples of the uterine cervix (double staining) Sections of formalin fixed, paraffin embedded tissue samples of the cervix uteri were immuno- fluorescent stained using antibodies specific for p16lNK4a and Ki67.

The tissue sections were rehydrated through incubation in xylene and graded ethanol, and transferred to Aqua bidest. Antigen Retrieval was earned out with 10mM citrate buffer (pH 6.0) for p16lNK4a and Ki67. Therefore the slides were heated in a waterbath for 40 min at 95 °C-98°C. The slides were cooled down to RT for 20 minutes, transferred to washing buffer (50mM Tris-HCI, 150mM NaCI, 0.05% Tween 20/DakoCytomation : code no.: S3006).

To avoid non-specific binding of the secondary antibody (species: goat) the specimens were incubated with 10% goat serum for 30min at RT.

The slides were then incubated with the primary antibodies, mouse anti-human p16lNK4a antibody (3. 48 pg/ml) and rabbit anti Ki67 (1: 25) for 30 min at RT, the slides were then rinsed with washing buffer and placed in a fresh buffer bath for 5 min. Excess buffer was tapped off and each specimen

was covered with 200NI of the secondary reagent, containing goat anti-mouse antibody, AlexaFluor@488 conjugated and goat anti rabbit antibody, Alexa Fluor546 conjugated and then incubated for 30min at RT. Then slides were washed two times as before and directly mounted with a special mounting medium for fluorescence.

The microscopic examination of the slides reveals, that cells immuno-reactive with p16lNK4a and being also immuno-reactive for Ki67 may only be found in samples, that may microscopically be identified as samples of dysplastic lesions. Cells stained by the p16lNK4a specific reaction, that are originating from metaplasias, are not stained by the reaction specific for Ki67. The microscopic inspection of the cell proliferation marker staining shows, that metaplastic cells over-expressing pl6lNK4a are not immuno-reactive with the antibodies directed against Ki67. Samples containing dysplastic tissue areas in contrast comprise cells, that are immuno-reactive with Kl67 and with antibodies directed against p16lNK4a. So in contrast to dysplasias in metaplasias no cells may be double stained using the Ki67 and p16lNK4a specific antibodies.

Figures 1-6 show the staining results for a severe dysplasia of the cervix uteri and for a squamous metaplasia using antibodies directed against Ki67 and those directed against p16lNK4a.

Immunoreactivity for Ki67 renders red fluorescence, immunoreactivity for p16lNK4a renders green fluorescence and overlay of red and green fluorescence renders yellow fluorescence. In the dysplastic specimen (Fig. 1-3) many cells of the dysplasia show nuclear positivity for Ki67 as well as positivity for p161NK4a and give thus rise to yellow fluorescence (see Figure 3) In contrast in the metaplastic specimen (Fig. 4-6) areas that are positively expressing p16lNK4a do not show any positive staining for Ki67 indicating lack of expression of the antigen in these areas. The double staining may be observed in this specimen (Fig. 6) and thus no cells in the specimen give rise to yellow fluorescence.

The results show, that the double staining of cells with reagents specific for Ki67 allows to discriminate p16lNK4a over-expressing metaplasias from dysplasias.

Example 2: Detection of cells co-expressing p14ARF and mcm2 in samples of the uterine cervix by in situ hybridization Smears of the uterine cervix may be semi-quantitatively analysed for the mRNA level of p16lNK4a and mcm2 in an in-situ staining reaction. The staining reaction is performed as follows : For rehydration the spray-fixed smears are incubated in fresh 50% EtOH on a rocking device. The PEG film produced by the fixation procedure is removed by intensive rinsing. Following the smears

are rinsed in aqua bidest. The smears are incubated with porteinase K (I Opg/ml in PBS) for 10 min at 37°C. Then the slides are transferred to washing buffer (PBS/0.1% Tween20) and finally the area containing the cells was surrounded with a lipid-pencil.

The hybridization mixture is prepared by mixing 50 pi of ready to use hybridization buffer (DAKO A/S, Glostrup, Danmark) with about 5-10 pmol of the probes. The probes are biotin-and Digoxygenin-labelled oligonucleotides of sequences complememtary to the respective mRNAs.

The hybridization mixture is heated to 95°C and afterwards equilibrated to 37°C. After the boiling procedure the smears are incubated with each 50 NI of the hybridization mixture for 4 hours at 42°C.

The samples are washed in excess volumes of the wash buffers two times in 2 x SSC at 37°C for 15 min and once in 1 x SSC at 37 °C for 15 min Then the smears are rinsed two times at room temperature in 2 x SSC. Following this washing procedure the dissections are incubated for 30 min with blocking buffer (NEN, Blockingbuffer) at room temperature. Then follows 1 hour incubation with a 1: 100 diluted (in Blocking buffer, see above) Strepavidine-alkaline phosphatase and with monoclonal mouse HRP-labeled anti-Digoxygenine antibodies (Molecular Probes). The smears are then washed 2 times in 1 x PBS/0, 1% Triton X-100 for 10 min at room temperature, followed by one wash step with 1 x PBS, 50 mM MgCl2 (pH 9,2) for 10 min at room temperature. Then the staining reaction is performed with ELF 97 phosphate (Molecular Probes) for 10 sec to 7 min at room temperature. Excess substrate is washed 3 times with 1x PBS/0, 1% Triton X-100 for 10 min at room temperature. In a second staining step, the section is incubated with Tyramides-Alexa-Fluor 594 for 10 sec to 7 min. Excess substrate is washed 3 times with 1x PBS/0, 1% Triton X-100 for 10 min at room temperature.. Finally the smears are dipped in H20dest. and embedded with Fluorescence mounting medium (DakoCytomation). Then the stained dissections can be analysed by fluorescence microscopy.

The microscopic examination of the slides reveals, that cells positive for expression of p16lNK4a and also expressing mcm2 may only be found in samples, that may microscopically be identified as samples of dysplastc lesions. Cells stained by the p16lNK4a specific reaction, that are identifiable as metaplasias, are not stained by the reaction specific for mcm2. The microscopic inspection of the mRNA hybridisabon shows, that metaplastic cells over-expressing p16lNK4a do not significantly express mRNA of mcm2. Dysplastc cells in contrast may be stained by in situ hybridisabon with probes specific for mcm2 and with probes directed against p16lNK4a. So in contrast to dysplastic cells in metaplastic cells no double staining using the Ki67 and p16lNK4a specific probes may be observed.

The results show, that the double staining of cells with reagents specific for mcm2 allows to discriminate p16 over-expressing metaplasias from dysplasias.

Example 3: Immuno-fluorescent detection of the over-expression of p1WNK4a and Ki42 in samples of the uterine cervix (double staining) Merckofix2) fixed cytological samples (conventional smears and liquid based cytology (ThinPreps@)) of the cervix uteri were immuno-fluorescent stained using antibodies specific for p16lNK4a and Ki-S2.

Conventional smears and liquid based cytological samples (ThinPrepsE)) were rehydrated in ethanol (50%) for 10min and transferred in Aqua bidest. Antigen Retrieval was carried out with 10mM citrate buffer (pH 6.0) for p16lNK4a and Ki67. Therefore the slides were heated in a waterbath for 40 min at 95 °C-98°C. The slides were cooled down to RT for 20 minutes, transferred to washing buffer (50mM Tris-HCI, 150mM NaCI, 0.05% Tween 20/DakoCytomation : code no.: S3006) and finally the samples were surrounded with a lipid-pencil.

To avoid non-specific binding of the secondary antibody (species: goat) the specimens were incubated with 10% goat serum for 30min at RT.

The slides were then incubated with the primary antibodies, mouse anti-human p16lNK4a antibody (clone E6H4) (3. 48 ug/ml) and rabbit anti Ki-S2 (1: 25) for 30 min at RT, the slides were then rinsed with washing buffer and placed in a fresh buffer bath for 5 min. Excess buffer was tapped off and the specimen was covered with 200pI of the secondary reagent, containing goat anti-mouse antibody, AlexaFluort4488 conjugated and goat anti rabbit antibody, Alexa Fluor@546 conjugated and then incubated for 30min at RT. Then slides were washed two times as before and directly mounted with a special mounting medium for fluorescence.

The microscopic examination of the slides reveals, that cells immuno-reactive with p16lNK4a and being also immuno-reactive for Ki-S2 may microscopically be identified as dysplastic cells. Cells stained by the p16"specific reaction, which are not stained by the reaction specific for Ki-S2, may be classified by an experienced pathologist as either being metaplasUc or being of endometrial origin. The microscopic inspection of the cell proliferation marker staining shows, that metaplastic cells over-expressing p16lNK4a are not immuno-reactive with the antibodies directed against Ki-S2.

Dysplastic cells in contrast are immuno-reactive with KI-S2 and with antibodies directed against p16lNK4a. So in contrast to dysplastic cells, in metaplasias no cells may be double stained using the Ki-S2 and p16NK4a specific anUbodies.

The results show, that the double staining of cells with reagents specific for Ki-S2 allows to discriminate p16lNK4a over-expressing metaplastic cells from dysplastic cells.

Example 4: Immuno-fluorescent detection of the over-expression of p16'NK4a, Ci67 and PCNA in bronchioalveolar-lavage samples of Individuals with diagnosed small cell lung cancer (double staining) Cells contained in bronchioalveolar lavage specimens of patients were prepared according to ThinPrep technology. MerokofixS fixed cytological samples of the lavages of patients with diagnosed small cell lung cancer were immuno-fluorescent stained using antibodies specific for p16'NK4a, Ki67and PCNA.

In this experiment a procedure was used, that does not discriminate between staining originating from immunoreactivity against the two proliferation markers PCNA and Ki67. Within the context of the present invention it must not necessarily be answered whether the one or the other proliferation marker is expressed in the cells. The main question is whether any proliferation marker is expressed.

Such in the course of the experiments a procedure was chosen, that gives the same fluorescence signal as indication of the presence of a proliferation marker. This procedure is suitable to improve sensitivity of the detection of the cell proliferation characteristics. As the case may be the procedure may be applied as to render one detectable signal for three four or even more marker molecules being characteristic for cell proliferation. Analogous the same may under certain circumstances be true for the INK4a gene expression products. It must be understood, that as the case may be different staining signals for different proliferation marker molecules may be desirable. The procedures may be applied to the necessities of the respective experiment.

The tissue sections were rehydrated through incubation in xylene and graded ethanol, and transferred to Aqua bidest. Conventional smears and liquid based cytological samples (ThinPrepsi)) were rehydrated in ethanol (50%) for 10min and transferred in Aqua bidest. Antigen Retrieval was carried out with 10mM citrate buffer (pH 6.0) for p16'NK4a, Ki67and PCNA. Therefore the slides were heated in a waterbath for 40 min at 95 °C-98°C. The slides were cooled down to RT for 20 minutes, transferred to washing buffer (50mM Tris-HCI, 150mM NaCI, 0.05% Tween 20/DakoCytomation : code no.: S3006) and finally the samples were surrounded with a lipid-pencil.

To avoid non-specific binding of the secondary antibody (species: goat) the specimens were incubated with 10% goat serum for 30min at RT.

The slides were then incubated with the primary antibodies, mouse anti-human pl6lNK4aantibody (3.48 pg/ml), rabbit anti Ki67 and rabbit anti PCNA (each 1: 25) for 30 min at RT, the slides were then rinsed with washing buffer and placed in a fresh buffer bath for 5 min. Excess buffer was tapped off and the specimen was covered with 200psi of the secondary reagent, containing goat anti-mouse antibody, AlexaFluor@488 conjugated and goat anti rabbit antibody, Alexa Fluor@546 conjugated and then incubated for 30min at RT. Then slides were washed two times as before and directly mounted with a special mounting medium for fluorescence.

The microscopic examination of the slides reveals, that cells immuno-reactive with p16'NK4a and being also immuno-reactive for Ki67 or PCNA may microscopically be identified as cells of small cell lung cancer. Cells stained by the p16'NK4a specific reaction, that are originating from metaplasias, are not stained by the reaction specific for Ki67 and PCNA. The microscopic inspection of the cell proliferation marker staining shows, that metaplastic cells over-expressing p16'NK4a are not immuno- reactive with the antibodies directed against Ki67 and PCNA. Samples containing dysplastic cells in contrast comprise cells, that are immuno-reactive with Ki67/PCNA and with antibodies directed against p161NK4a. So in contrast to dysplasias in metaplasias no cells may be triple stained using the Ki67 and PCNA and p16'NK4a specific antibodies.

The results show, that the triple staining of cells with reagents specific for Ki67/PCNA allows to discriminate p16'NK"e over-expressing non-dysplastic cells from dysplasias.

Example 5: Flow cytometric detection of dysplastic cells by simuitaneous detection of mcm5 mRNA, p14ARF protein and Ki67 protein in cells of cervical origin Cytological samples (Cell suspensions in PBS, pH 7,4) of the cervix uteri were fluorescent stained using antibodies specific for p14ARF and Ki-67 and oligoprobes for mcm-5 and evaluated by three- color fluorescent FACS analysis.

Cells were centrifuged, the supematant decanted and fixed and permeabilized with 100 ml Permeafix (Ortho Diagnostic, Raitan, NJ, USA) 1 hour at room temperature. Cells were washed in sterile PBS, pH 7,4, pelleted and resuspended in 100 ml Permeafix 1 hour at room temperature.

Cells were washed in sterile PBS, pH 7,4, pelleted and resuspended in sterile PBS. They were incubated with PE-conjugated anti-p14ARF antibody and PE-Cy5-conjugated anti-Ki67 antibody for 1 h at +4°C. Cells were washed in sterile PBS, pH 7,4, pelleted and resuspended in 100 ml Permeafix 30 min at room temperature. Cells were washed in sterile PBS, pelleted by centrifugation, and then washed again in 2x standard saline citrate (SSC). After centrifugation, the cell pellet was

resuspended in hybridisation solution (2x SSC, 30% formamide, sonicated salmon sperm, yeast transfer DNA) containing 500 ng of 5-carboxy-fluorescein double end labelled, mcm5-specific oligonucleotides probes. The intercellular hybridisation was performed at 42°C for 1 hour, followed by successive washes in 2x SSC, 0,5 % Triton X-100, and 1x SSC, 0,5 % Triton X-100 at 42°C. The cells were resuspended for analysis in PBS, pH 8,3 and analysed on a flow cytometer (FACScan, Becton Dickinson, IS). For each analysis 30.000-100. 000 gated events were collected. Data analysis was performed using CeIlQuest (Becton Dickinson, IS) The flow cytometer analysis reveals, that cells immuno-reactive with p14ARF and being also immuno-reactive for Ki-67 and/or reactive for the oligoprobes of mcm5 could be only identified in samples of patients with dysplastic lesions of the cervics. Samples from women with no dysplasbc lesions showed no concomitant staining of p14ARF with Ki67 or mcm5. The results show, that the double or triple staining of cells with reagents specific for Ki67 and/or mcm5 allows to discriminate p14ARF over-expressing non-dysplastic cells from p14ARF over-expressing dysplastic cells.

Example 6: Immunoenymatic detection of the over-expression of p161NK4a and Ki67in histological samples of the uterine cervix (sequential double staining) Sections of formalin fixed, paraffin embedded tissue samples of the cervix uteri were immunoenzymatic double stained using antibodies specific for p161NK4a and Ki67.

The tissue sections were rehydrated through incubation in xylene and graded ethanol, and transferred to Aqua bidest. Antigen Retrieval was earned out with 10mM citrate buffer (pH 6.0) for p161NK4a and Ki67. Therefore the slides were heated in a waterbath for 40 min at 95 °C-98°C.

The slides were cooled down to RT for 20 minutes and transferred to washing buffer ( DakoCytomation).

Endogenous peroxidase activities were blocked with 3% H202 (DakoCytomation) for 5min at RT.

After washing the slides for 5min at RT they were incubated with the first primary antibody, mouse anti-human p161NK4a antibody (MTM) for 30 min at RT and were then rinsed with washing buffer and placed in a fresh buffer bath for 5 min. Excess buffer was tapped off and each specimen was covered with 200pI of the secondary reagent (EnVision goat anti mouse-Peroxidase/ DakoCytomation) and incubated for 30min at RT. Then slides were washed three times as before. For the chromogenic visualization DAB (DakoCytomation) was used by incubating the slides

with the substrate chromogen complex for 10min at RT. The reaction was stopped in deionized water and slides placed in wash buffer.

After washing the slides were incubated with the second primary antibody, rabbit anti-human Ki67 antibody (Dianova, clone Ab-3) for 30 min at RT and were then rinsed with washing buffer and placed in a fresh buffer bath for 5 min. Excess buffer was tapped off and each specimen was covered with 200pI of the secondary reagent (goat anti rabbit-Alkaline Phosphatase labeled/ DakoCytomation) and incubated for 30min at RT. Then slides were washed three times as before.

For the chromogenic visualization FastRed (BioGenex) was used by incubating the slides with the substrate chromogen complex for 30min at RT. The reaction was stopped in deionized water.

After counterstaining with hematoxylin (DakoCytomation) for 2min at RT the slides were incubated in running tap water for 10min at RT and then mounted with aqueous mounting medium (Aquatex/ MERCK).

The microscopic examination of the slides reveals, that cells immuno-reactive with p161NK4a and being also immuno-reactive for Ki67 may only be found in samples, that may microscopically be identified as samples of dysplastic lesions. Cells stained by the p161NK4a specific reaction, that are originating from metaplasias, are not stained by the reaction specific for Ki67. The microscopic inspection of the cell proliferation marker staining shows, that metaplastic cells over-expressing p161NK4a are not immuno-reactive with the antibodies directed against Ki67. Samples containing dysplastic tissue areas in contrast comprise cells, that are immuno-reactive with Ki67 and with antibodies directed against p161NK4a. So in contrast to dysplasias in metaplasias no cells may be double stained using the Ki67 and p161NK4a specific antibodies.

The results as given in Figure 7 show, that the double staining of cells with reagents specific for Ki67 and p161NK4a allows for a specific double staining pattern also in chromogenic staining procedures.

Example 7 : immunoenzymatic detection of the over-expression of pi and Ki67in cytological samples of the uterine cervix (sequential double staining) Merckofixi) fixed cytological samples (conventional smears and liquid based cytology (ThinPrepsi))) of the cervix uteri were immunoenzymatic double stained using antibodies specific for p16lNK4a and Ki67.

Conventional smears and liquid based cytological samples were rehydrated in ethanol (50%) for 10min at RT and transferred in Aqua bidest. Antigen Retrieval was carried out with 10mM citrate buffer (pH 6.0) for p16lNK4a and Ki67. Therefore the slides were heated in a waterbath for 40 min at 95 °C-98°C. The slides were cooled down to RT for 20 minutes and transferred to washing buffer.

Endogenous peroxidase activities were blocked with 3% H202 for 5min at RT.

After washing the slides they were incubated with the first primary antibody, mouse anti-human p16lNK4a antibody for 30 min at RT and were then rinsed with washing buffer and placed in a fresh buffer bath for 5 min. Excess buffer was tapped off and each specimen was covered with 200pI of the secondary reagent (EnVision goat anti mouse-Peroxidase) and incubated for 30min at RT.

Then slides were washed three times as before. For the chromogenic visualization DAB ( DakoCytomation) was used by incubating the slides with the substrate chromogen complex for 10min at RT. The reaction was stopped in deionized water and slides placed in wash buffer.

After washing the slides were incubated with the second primary antibody, rabbit anti-human Ki67 antibody (Dianova, clone Ab-3) for 30 min at RT and were then rinsed with washing buffer and placed in a fresh buffer bath for 5 min. Excess buffer was tapped off and each specimen was covered with 200NI of the secondary reagent (goat anti rabbit-Alkaline Phosphatase labeled/ DakoCytomation) and incubated for 30min at RT. Then slides were washed three times as before.

For the chromogenic visualization Fast Red (BioGenex) was used by incubating the slides with the substrate chromogen complex for 30min at RT. The reaction was stopped in deionized water.

After counterstaining with hematoxylin (DakoCytomation) for 2min at RT the slides were incubated in running tap water for 10min at RT and then mounted with aqueous mounting medium (Aquatex I MERCK).

The microscopic examination of the slides reveals, that cells immuno-reactive with p16lNK4a and being also immuno-reactive for Ki67 may be identified as dysplastic cells on the basis of their morphology. Cells stained by the p16lNK4a specific reaction, that are originating from metaplasias, are not stained by the reaction specific for Ki67. The microscopic inspection of the cell proliferation marker staining shows, that metaplastic cells over-expressing p16lNK4a are not immuno-reactive with the antibodies directed against Ki67. Dysplastc cells in contrast are immuno-reactive with Kl67 and with antibodies directed against p16lNK4a. So in contrast to metaplastc cells in dysplastic cells double-staining of single cells may be produced using the Ki67 and p16lNK4a specific antibodies.

The results as given in Figure 8 show, that the double staining of cells with reagents specific for Ki67 and p161NK4a allows for a specific double staining pattern also in chromogenic staining procedures.