WO2013041731A1

WO2013041731A1 - Marker gene based diagnosis, staging and prognosis of prostate cancer

Info

Publication number: WO2013041731A1
Application number: PCT/EP2012/068810
Authority: WO
Inventors: Kyrylo Litovkin; Aleyde Van Eynde; Mathieu Bollen; Monique Beullens
Original assignee: Katholieke Universiteit Leuven
Priority date: 2011-09-23
Filing date: 2012-09-24
Publication date: 2013-03-28
Also published as: GB201206209D0; WO2013149904A1

Abstract

75 Abstract This invention relates generally to a method of diagnosis and prognosis, in particular staging and/or typing and/or predicting outcome, for distinguishing between a benign prostate hyperplasia and a prostate cancer and between an hormone sensitive and an hormone refractory prostate cancer condition and specifically to identification of differentially methylated CpG islands in the regulatory regions surrounding the transcriptional start site of at least one marker gene of the present invention as a diagnostic and/or prognostic indicator of prostate cancer (PrCa) and for distinguishing androgen-refractory from androgen sensitive prostate cancer. The marker genes of the present invention comprise TDRD1, RARB, GSTP1, APC, CCND2 PTGS2, BCL2, RASSF1, LGALS3, CDH13, PITX2, and HOXD3. This invention relates more specifically to the detection of hypomethylation of said regulatory region of the Marker gene TDRD1 together with the hypermethylation of at least one marker gene, selected from the list: RARB, GSTP1, APC, CCND2 PTGS2, BCL2, RASSF1, LGALS3, CDH13, PITX2, and HOXD3. Additional hypomethylation markers of this invention comprise MAGEA2 and/or MAGEA11), and additional hypermethylation markers of this invention comprise TDRD5, TBX20, SOX1 and/or MSMB, with MSMB being hypermethylated in non-CpG islands. This invention further relates to the prediction, prognosis or diagnosis of prostate cancer, including metastasis, more particularly in patients with prostate cancer. Marker genes have been identified of which promoter regions containing differentially methylated regions, compared to a reference sample, which are indicative for the prediction or prognosis of prostate cancer.

Description

Marker gene based diagnosis, staging and prognosis of prostate cancer

FIELD OF THE INVENTION

This invention relates generally to a method of diagnosis and prognosis, in particular staging and/or typing and/or predicting outcome, for distinguishing between a benign prostate hyperplasia and a prostate cancer and between an hormone sensitive and an hormone refractory prostate cancer condition and specifically to identification of differentially methylated CpG islands in the regulatory regions surrounding the transcriptional start site of at least one marker gene of the present invention as a diagnostic and/or prognostic indicator of prostate cancer (PrCa) and for distinguishing androgen-refractory from androgen sensitive prostate cancer.

The marker genes of the present invention comprise TDRDl, RARB, GSTP1, APC, CCND2 PTGS2, BCL2, RASSF1, LGALS3, CDH13, PITX2, and HOXD3. This invention relates more specifically to the detection of hypomethylation of said regulatory region of the Marker gene TDRDl together with the hypermethylation of at least one marker gene, selected from the list: RARB, GSTP1, APC, CCND2 PTGS2, BCL2, RASSF1, LGALS3, CDH13, PITX2, and HOXD3. Additional hypomethylation markers of this invention comprise MAGEA2 and/or MAGEA11), and additional hypermethylation markers of this invention comprise TDRD5, TBX20, SOXl and/or MSMB, with MSMB being hypermethylated in non-CpG islands.

This invention further relates to the prediction, prognosis or diagnosis of prostate cancer, including metastasis, more particularly in patients with prostate cancer. Marker genes have been identified of which promoter regions containing differentially methylated regions, compared to a reference sample, which are indicative for the prediction or prognosis of prostate cancer.

BACKGROUND OF THE INVENTION

Prostate cancer (PrCa), is the second most common malignancy in males worldwide after lung cancer, and the third leading cause of cancer death in men. Early detection greatly improves survival rates. If the malignant prostate tumor is still local, PrCa can be successfully treated by radiation therapy, surgery, hormone therapy and/or chemotherapy. Unfortunately, if the PrCa invades other parts of the body like bones, lymph nodes, rectum and bladder (metastatic PrCa), it becomes refractory to hormone therapy. For this advanced PrCa the prognosis is poor. Currently, PrCa is detected by an elevated level of Prostate- Specific Antigen (PSA) in the blood, along with a digital rectal exam. The PSA test is also used to monitor patients for the recurrence of PrCa following surgery or other treatments. However, although the PSA test has greatly improved the detection of PrCa, its usefulness is still controversial. A recent study by Concato et al. shows that PSA screening is not associated with lower mortality (Concato J, et al. (2006) Arch Intern Med. 166:38-43). Moreover, the serum PSA level is also elevated in non-cancerous prostate disorders such as benign prostate hyperplasia and infection.

Initial tests for suspected prostate cancer is done by analysis of blood levels of proteins like PSA or for instance PSP94 protein. Positive tests are followed by a conformational diagnosis. The only test which can fully confirm the diagnosis of prostate cancer is a biopsy, the removal of small pieces of the prostate for microscopic examination. The present invention provides a novel diagnostic test of prostatic tissue or cells obtainable from prostatic tissue.

A condition of benign prostatic hyperplasia (BPH), or benign prostatic hypertrophy is common as a man ages. It is thus very important to distinguish between a PrCa and a BPH. Moreover hormone-refractory prostate cancers are more aggressive and need specific treatments such as apoptosis and regression induction of the tumors and/or antimetastasis. Besides or in addition to the presently used prostate cancer nomograms, there is thus a need in the art for additional prostate cancer screening or diagnosis methods, and more particularly for a biomarker(s) that can discriminate between benign and malignant tumors and between aggressive and indolent (slow-growing) cancers. The present invention fulfills these needs.

The latter type of cancer will remain localized in a person's lifetime and is unlikely to reduce life expectancy. In contrast, an aggressive cancer is more lethal, due to metastasis, and requires immediate intervention. Therefore, there is an unmet need for a reliable diagnostic assay and biomarker(s) to distinguish between these two types.

We have now found that the methylation status of the regulatory region surrounding the transcription start site (TSS) of TDRDl (tudor domain-containing protein 1) is indicative of the type and/or stage of a prostate cell proliferative disorder, and can thus be considered to be a biomarker(s) that can discriminate between different types and/or stages of prostate cancer. TDRD1 was discovered in a systematic search for genes expressed in mouse spermatogonia but not in somatic tissues by Wang et al. in 2001 (Wang, P. J., et al. Nature Genet. 27: 422- 426, 2001). Loriot et al, 2003 revealed that TDRD1 expression is upregulated in prostate vs normal samples in 10 out of 26 prostate cancer patients. Furthermore, in vitro treatment of melanoma cell lines, sarcoma cell lines or PHA-activated peripheral blood lymphocytes with a demethylating agent was shown to induce TDRD1 expression. Loriot et al., 2003 is however silent about the methylation status of TDRD1 in vivo in prostate cancer patients, let it be to suggest a possible prognostic value of said methylation status. Furthermore, contrary to other known prognostic markers of prostate cancer such as PITX2 (Weiss et al, J. Urol; 181(4) 1678-1685; 2009) or HOXD3 (Kron et al, Laboratory Investigation; 90. 1060-1067; 2010), which are hypermethylated in tumors, we have found that TDRD1 is hypomethylated in tumor versus normal tissue.

SUMMARY OF THE INVENTION

The present invention solves the problems of the related art by providing a diagnostic and prognostic assay that allows one to determine the predisposition to, or the incidence of prostate cancer and allows to distinguish between different types and/or stages of cancer, in particular between hormone-refractory and hormones-sensitive cancer, particularly in prostatic tissues or cells originating from prostatic tissues and to predict outcome. However, the test could also be used on body fluids.

In accordance with the purpose of the invention, as embodied and broadly described herein, the invention is broadly drawn to methods and assays for detecting a prostate proliferative disorder, in particular for identifying prostate tumor cells that have become refractory or resistant to hormone therapy, and thus allowing to identify the prostate cancer or/and to distinguish hormone sensitive from hormone refractory prostate cancers.

The present invention relates generally to the identification of the distinguishing difference between a hormone refractory prostate tissue cellular proliferative disorder and a hormone sensitive prostate tissue cellular proliferative disorder in a subject, preferably a human subject. The distinguishing difference relies on the identification of one or more hypomethylated CpG islands surrounding the transcription start site (TSS) of the human gene TDRD1 either or not in combination with hypermethylation of CpG islands surrounding the transcription start site (TSS) of at least one gene selected from RARB, GSTP1, APC, CCND2, PTGS2, BCL2, RASSF1, LGALS3, CDH13, PITX2, and HOXD3, more in particular the hypo- and hypermethylated CpG islands are found in regions upstream of the TSS or in the promoter region of said human genes. In certain embodiments of this invention, the method further comprises the analysis of (hyper)methylation in the MSMB gene, more particularly the non-CpG dinucleotides in the regulatory region surrounding the TSS of said MSMB gene, wherein the hypermethylation of non-CpG dinucleotides in said region is indicative of a predisposition to, or the incidence of, prostate cancer.

In particular embodiments of the present invention, the analysis of the methylation status of the genes is restricted to at least one gene, more preferably at least two genes, or three or four genes selected from the group TDRD1, PITX2, RASSF1, and HOXD3. In particular embodiments said group of genes (TDRD1, PITX2, RASSF1, and HOXD3) are to be analysed for their methylation status, and are used to predict the incidence of and more particular the aggressiveness of prostate cancer. These four markers: TDRD1, PITX2, RASSFl, and HOXD3, are thus particularly useful as prognostic markers, more specifically in the current invention TDRD1 hypomethylation and/or hypermethylation of at least one, two or three genes from the group of PITX2, RASSFl, and HOXD3 are indicative for a negative prognosis, or an indication for an aggressive tumor, more particular a prostate tumor. In a certain embodiment of this invention, this set of four genes (TDRD1, PITX2, RASSFl, and HOXD3) or any combination of two or three of these four genes can be used in a method of the invention to decide on the proper treatment or proper medicament of the patient. In a more particular embodiment, the method of the present invention wherein TDRDl hypomethylation and/or hypermethylation of at least one, two or three genes from the group of PITX2, RASSFl, and HOXD3 is detected in a sample of prostatic tissue or in a biological sample that comprises prostatic cells or prostatic cell components from a human patient when comparing the methylation level of said genes in a reference sample, is used to decide on the proper treatment of said patient, in particular the methylation level of said genes is indicative for the decision about the initiation or continuation of a proper treatment, wherein in a more particular embodiment said proper treatment is selected from a prostatectomy, treatment with a methylation inhibitor, treatment with a compound which reduces male hormones, radiotherapy, or treatment with neutraceuticals. In certain embodiments of the present invention, the analysis of the methylation status is restricted to the analysis of the methylation status of the TDRD1 gene, more particularly hypomethylation of said TDRD1 gene detected in a sample of prostatic tissue or in a biological sample that comprises prostatic cells or prostatic cell components from a human patient when comparing the methylation level of said TDRD1 gene in a reference sample, is indicative for prostate cancer or indicative for the predisposition to prostate cancer, more particularly, an aggressive or high grade prostate cancer.

The present invention comprises all combinations of the four prognostic markers: TDRD1, PITX2, RASSFl and HOXD3, ie. any combination of two or three or all four of these markers can be used in the methods of the present invention. With regard to the four prognostic markers of this invention, TDRD1 is a hypomethylation marker and PITX2, RASSFl, and HOXD3 are hypermethylation markers, meaning that hypermethylation of PITX2 and/or RASSFland/or HOXD3, and/or hypomethylation of TDRD1 when comparing the methylation status of a patient or a human being suspected to have prostate cancer, to the methylation level of said genes in a reference sample is indicative for prostate cancer or indicative for the predisposition to prostate cancer, more particularly for typing and/or staging tumors, in particular to identify an aggressive or high grade prostate cancer and for predicting outcome, in particular to predict biochemical recurrence (BCR) and/or clinical failure. A particular combination of said prognostic markers that can be used in the methods of the present invention is the combination of TDRD1, with at least one marker selected from the group of RASSFl, PITX2 and HOXD3. This comprises the combination PITX2 and TDRD1; the combination TDRD1 and RASSFl; the combination of TDRD1 and HOXD3; the combination PITX2, TDRD1, and RASSFl; the combination PITX2, TDRD1 and HOXD3; and the combination TDRD1, RASSFl, and HOXD3; the combination TDRD1, PITX2, HOXD3, and RASSFl . Other particular combinations of prognostic markers that can be used in the methods of the present invention is the combination of PITX2, with at least one marker selected from the group of TDRD1, RASSFl and HOXD3. This comprises the combination PITX2 and TDRD1; the combination PITX2 and RASSFl; the combination of PITX2 and HOXD3; the combination PITX2, TDRD1, and RASSFl; the combination PITX2, TDRD1 and HOXD3; and the combination PITX2, RASSFl, and HOXD3. Other particular combinations of prognostic markers that can be used in the methods of the present invention are the combination of RASSFl, with at least one marker selected from the group of PITX2, TDRD1 and HOXD3, this comprising the combination RASSFl and TDRD1; the combination RASSF1 and HOXD3; the combination RASSF1 and PITX2; the combination RASSF1, TDRD1 and HOXD3; and the combination RASSF1, PITX2 and TDRD1. Other particular combinations of prognostic markers that can be used in the methods of the present invention are the combination of HOXD3, with at least one marker selected from the group of PITX2, TDRD1 and RASSF1, this comprising the combination HOXD3 and TDRD1; the combination HOXD3 and PITX2; the combination HOXD3 and RASSF1; the combination HOXD3, RASSF1, and TDRD1; and the combination HOXD3, PITX2 and TDRD1.

The present invention also comprises the hypomethylation markers MAGEA2 and/or MAGEAl l). These hypomethylation markers can be analysed separately or in combination with the hypomethylation marker TDRD1, the use and analysis of such hypomethylation marker is as described for TDRD1 in the present invention.

The present invention also comprises additional hypermethylation markers such as TDRD5, TBX20, SOX1 and/or MSMB, with MSMB being hypermethylated in non-CpG islands. The use and analysis of such hypermethylation markers is as described for the other methylation markers of the present invention (eg. RARB), with the exception for the MSMB gene wherein the methylation is analysed on non-CpG islands, as described in the present invention.

The prognostic methods that detect whether a prostate cancer in subjects, preferably human, comprises an androgen refractory cancer and/or an androgen sensitive cancer can be carried out by analysis of the methylation status of said genes in a sample of a subject. Thus, in a first aspect the invention provides methods for detecting, and in particular for typing and/or staging and/or prediction of outcome; in a subject a prostate cell proliferative disorder, which methods comprise the steps of:

a) obtaining a biological sample from the subject;

b) determining the methylation state of CpG island(s) upstream and/or downstream of the TSS and/or in the promoter region of the TDRD1 gene either or not in combination with determining the methylation state of CpG island(s) upstream and/or downstream of the TSS and/or in the promoter region of at least one gene selected from RARB, GSTP1, APC, CCND2, PTGS2, BCL2, RASSF1, LGALS3, CDH13, PITX2, and HOXD3, and wherein detection of hypomethylation in these region(s) of TDRD1 and/or hypermethylation of at least one, two, three, four or five genes selected from the group of genes: RARB, GSTP1, APC, CCND2, PTGS2, BCL2, RASSFl, LGALS3, CDH13, PITX2, and HOXD3 is indicative of the type and/or stage, the predisposition to, or the incidence of prostate cancer.

In a similar aspect, the invention provides methods for detecting in a subject an androgen refractory prostate cancer, which methods comprise the steps of:

a) obtaining a biological sample from the subject;

b) determining the methylation state of CpG island upstream and/or downstream of the TSS region and/or in the promoter region of at least one, two, three, four or five genes selected from the panel of genes: TDRD1, RARB, GSTP1, APC, CCND2, PTGS2, BCL2, RASSFl, LGALS3, CDH13, PITX2, and HOXD3, in the subject's sample; and wherein detection of hypomethylation of TDRD1 and/or hypermethylation of one or more of the other gene(s) is indicative of a predisposition to, or the incidence of, androgen sensitive prostate cancer.

Preferably, both the methods of the invention comprise a further step as follows: c) identifying methylation of region(s), wherein hypomethylation of TDRD1 and/or hypermethylation of at least one, two, three or four genes of the group: TDRD1, RARB, GSTP1, APC, CCND2, PTGS2, BCL2, RASSFl, LGALS3, CDH13, PITX2, and HOXD3 is identified as being different when compared to the same region(s) of the gene or associated regulatory region in a subject having an androgen sensitive prostate cancer. Another aspect of the invention is that it provides methylation conditions of regulatory regions of the panel of genes (TDRD1, RARB, GSTP1, APC, CCND2, PTGS2, BCL2, RASSFl, LGALS3, CDH13, PITX2, and HOXD3), such as in the CpG islands surrounding the TSS of said panel of human genes, which can be used (a) to analyze the presence of cancer cells in prostate tissue and/or in prostatic secretions, for instance in seminal plasma and (b) to define patients that have a prostate cancer or alternatively patients that have a normal prostate, and (c) to define which patients with a prostate cancer have an androgen refractory prostate cancer or alternatively to define which patients with a prostate cancer have an hormone sensitive prostate cancer.

Such test provides an accurate means or tool to decide about the suitable treatment of the prostate cancer; in particular if the TDRD1 gene is hypomethylated and/or at least one, two,three, four, or five other genes of said panel of genes is/are methylated/hypermethylated the need for chemotherapy, surgery or radiation therapy is identified. The methods of present invention can also be used to predict effectiveness of such chemotherapies applicable on a prostate cancer.

Patients affected by a condition of hypermethylation of regulatory regions of the genes of the group: RARB, GSTPl, APC, CCND2, PTGS2, BCL2, RASSF1, LGALS3, CDH13, PITX2, HOXD3 and MSMB such as in the CpG islands surrounding said genes (for MSMB hypermethylation is analyzed in non-CpG islands), and/or CpG islands upstream of the TSS or in the promoter region of said genes can for instance be treated by DNA methyltransferase (DNMT) inhibitors or can be treated with inhibitors of the EZH2 gene expression or inhibitors of the function of the polycomb protein EZH2 to induce a repair of abnormal methylation. In certain embodiments of the present invention, the analysis of the methylation status of the panel of genes is restricted to the genes RARB, GSTPl, APC, CCND2, and PTGS2. In certain embodiments of the present invention, hypermethylation of at least one, two, three or four genes of the group RARB, GSTPl, APC, CCND2, and PTGS2, is indicative of the incidence of prostate cancer. A preferred group of diagnostic markers, more particular prostate cancer diagnostic markers, identified in the present invention consists of RARB, GSTPl, APC, CCND2, and PTGS2. A particular combination of said diagnostic markers that can be used in the methods of the present invention, particularly in the diagnostic methods of the present invention, is the combination of all five diagnostic markers: RARB, GSTPl, APC, CCND2, and PTGS2, more in particular APC, GSTPl and RARB. This invention further comprises all different combinations of at least two, three or four genes of the group RARB, GSTPl, APC, CCND2, and PTGS2 which can be used in the methods of the present invention, particularly in the diagnostic methods of the present invention.

Still another aspect of the invention relates to the observation that due to the fact that the MSMB gene, which encodes PSP94 (beta-microsemenoprotein or beta-inhibin), a prostatic secretory protein of 94 amino acids, or PSP57 (lacking an internal exon of 106 bases in the coding region resulting in a frameshift at the 3' end, compared to PSP94 ) is repressed in hormone-refractory cancer cells, by the hypermethylation of a CpG island in the regulatory regions surrounding the transcriptional start site of the MSMB gene or in the promoter region that the encoding by the MSMB gene or expression of PSP94, known to be a suppressor of tumor growth and metastasis and to be secreted by the prostate gland and functions, is lost in advanced hormone-refractory cancer, for instance advanced hormone-refractory prostate cancer. PSP57 mRNA is in prostate tumor cell lines, aberrantly spliced and localized in the nuclear fraction of the cell. [Xuan JW, et al. Oncogene. 1995 Sep 21;l l(6): 1041-7. PSP57 mRNA has been also detected in other urogenital tissues (kidney, bladder) and in most tumor cell lines tested, but was not detectable in other tissues such as breast and lung. [Hoffmann, Pv., et al;. Nature Genetics 36, 664 (2004)"] Hypermethylation can be detected by restriction endonuclease treatment and Southern blot analysis. Therefore, in a method of the invention, when the cellular component detected is DNA, restriction endonuclease analysis is preferable to detect hypermethylation of the regulatory region of the selected genes of the invention, in their promoter or upstream of their promoter. Any restriction endonuclease that includes CG as part of its recognition site and that is inhibited when the C is methylated, can be utilized. Preferably, the methylation sensitive restriction endonuclease is BssHII, Mspl, or Hpall, used alone or in combination. Other methylation sensitive restriction endonucleases will be known to those of skill in the art.

Diagnosis of hypermethylation of the CpG island in the regions surrounding the TSS or in the promoter of certain genes of this invention (RARB, GSTP1, APC, CCND2, PTGS2, BCL2, RASSFl, LGALS3, CDH13, PITX2, HOXD3 and MSMB, wherein for MSMS hypermethylation is analyzed in non-CpG islands) can thus be used as a decision tool for treatment of a patient affected with such hypermethylation with a therapeutically effective amount of an DNA methyltransferase (DNMT) inhibitor for treating the prostate cancer or for preventing that a androgen sensible prostate cancer evolves into an androgen refractory prostate cancer. MGI Pharma developed small molecule DNA methyltransferase (DNMT) inhibitors for the treatment of cancer. Short oligonucleotide DNA methylation inhibitors in the art are Decitabine 5-Aza-CdR, SI 10 AzapG, S53 GpAza, S54 GpAzapG, S55 AzapGpAzapG, S56 pGpAzapAzapG, S52R AzapsG, Zebularine and SI 12 HEGpAzapG. A specific DNMT inhibitor is for instance the compound called SI 10 of the company SuperGen which is a dinucleotide containing decitabine, S 110, which has superior activity due to increased stability because of less degradation by hydrolytic cleavage and deamination. This is a DNA demethylating agent with a similar activity as decitabine (5-aza-2'-deoxycytidine) or its derivatives. Decitabine is a potent DNA methylation inhibitor which is approved in the US for the treatment of myelodysplasia syndromes (Yoo DB, et al. Cancer Research. 67: 6400- 6408, No. 13, 1 Jul 2007). Another DNA methyltransferase (DNMT) inhibitor is MG 98 (HYB 101584) is described in US 6953783 and US 6506735. MG 98 is a second generation antisense oligonucleotide that selectively targets DNA methyltransferase 1 (DNMT1) mRNA. By inhibiting the production of DNMT, the methylation of DNA is reversed and leads to re- expression of the tumour suppression genes. MG 98 is created by MethylGene Inc. (Stewart D, et al. 11th NCI-EORTC-AACR symposium on new drugs in cancer therapy. : 148, 7 Nov 2000. ; Winquist E, et al. European Journal of Cancer. 38 (Suppl. 7): 141, Nov 2002. ; Stewart DJ, et al. Annals of Oncology. 14: 766-774, May 2003 and Ramchandani S, et al. Proceedings of the National Academy of Sciences of the United States of America. 94: 684- 689, Jan 1997. and Davis AJ, et al. 11th NCI-EORTC-AACR symposium on new drugs in cancer therapy. : 94, 7 Nov 2000. These compounds can be administered in a therapeutically efficient amount to patients that have been identified by the diagnostic method of present invention to be in need thereof.

Thus, epigenetic loss of gene function due to hypermethylation can be rescued by the use of DNA demethylating agents and/or DNA methyltransferase inhibitors and/or HDAC inhibitors. Accordingly, the invention also provides for a method for predicting the likelihood of successful treatment of prostate proliferative disorder or prostate cancer, with a DNA demethylating agent and/or a DNA methyltransferase inhibitor and/or HDAC inhibitor comprising detecting a methylation change in the region surrounding the TSS or the promoter region of certain genes of this invention (RARB, GSTP1, APC, CCND2, PTGS2, BCL2, RASSF1, LGALS3, CDH13, PITX2, HOXD3 and MSMB) wherein detection of the methylation change is indicative of successful treatment to a higher degree than if the methylation modification is not detected.

The method of the present invention is also very suitable for identifying patients in which the prostate cell proliferative disorder is of a methylation-independent type. These types of cancers are not caused by changes in methylation of the involved genes and thus treatment with demethylation agents is not suitable for this group of patients. Patients having methylation-independent type of prostate cell proliferative disorder are characterized by iso- methylation of CpG dinucleotides in the regulatory region surrounding the TSS of TDRDl in combination with iso-methylation or hypomethylation of said region in at least one, more in particular at least 2, 3, 4, 5, 6, 7, 8, 9, 10 or 11 genes selected from RARB, GSTP1, APC, CCND2, PTGS2, BCL2, RASSF1, LGALS3, CDH13, PITX2, and HOXD3; in particular said genes are selected from the list comprising GSTP1, RARB, PITX2 and HOXD3 or in particular RASSF1, LGALS3, and CDH13. In a particular embodiment, TDRDl is not differentially methylated and RARB, GSTP1, APC, CCND2, PTGS2, BCL2, RASSFl, LGALS3, CDH13, PITX2, and HOXD3 are not differentially methylated or hypomethylated in methylation-independent prostate cancers as compared to benign prostatic tissue. Also provided is a kit for typing and/or staging and/or predicting outcome, detecting a predisposition to, or detecting the incidence of, prostate cancer in a sample comprising:

(a) means for detecting a methylation change in the region surrounding the TSS or the promoter region of at least one, two, three, four, five or six genes of the panel of genes: TDRD1, RARB, GSTP1, APC, CCND2, PTGS2, BCL2, RASSFl, LGALS3, CDH13, PITX2, and HOXD3.

(b) means for processing a sample derived from the prostate.

In particular embodiments of this invention, said panel of genes is restricted to the group of genes: RARB, GSTP1, APC, CCND2, PTGS2, BCL2, RASSFl, LGALS3, CDH13, PITX2, and HOXD3. In other embodiments of this invention, the panel of genes to be analysed is the analysis of the methylation status of the TDRD1 gene together with the analysis of the methylation status of at least one gene selected from the group: RARB, GSTP1, APC, CCND2, PTGS2, BCL2, RASSFl, LGALS3, CDH13, PITX2, and HOXD.

In particular embodiments the analysis of said genes is restricted to the region surrounding their TSS. In particular embodiments said region extends from 1.5 kb upstream to about 1.5 kb downstream from the transcription start site of said genes. In other particular embodiments, said region extends from 1.0 kb upstream to about 1.0 kb downstream from the transcription start site of said genes. In more particular embodiments of this invention the detection of hypomethylation in said region of the TDRD1 gene and/or the hypermethylation in said region of at least one gene selected from the group: RARB, GSTP1, APC, CCND2, PTGS2, BCL2, RASSFl, LGALS3, CDH13, PITX2, and HOXD3 indicates the presence of prostate cancer cells or is indicative of a predisposition to, or the incidence of, prostate cancer. In a particular embodiment of this invention, said hyper- and/or hypomethylation is detected when comparing the methylation status of the DNA of a test sample to the methylation status of a control sample and/or a benign prostate hyperplasia sample.

In particular embodiments of this invention, detection or analysis of hypomethylation in the TDRD1 gene further comprises detection or analysis of hypomethylation in the MAGEA2 and/or MAGEA11 gene. In particular embodiments of the present invention, detection or analysis of hypermethylation in at least one gene selected from the group: RARB, GSTP1, APC, CCND2, PTGS2, BCL2, RASSFl, LGALS3, CDH13, PITX2, and HOXD3, further comprises detection or analysis of hypomethylation in the SOX1 gene and/or the hypermethylation of non-CpG islands of the MSMB gene.

In the context of the present invention, hypermethylation and/or hypomethylation of the (marker) genes of this invention has the meaning of differential methylation i.e. hypermethylation (increased) and/or hypomethylation (decreased) of said genes, when compared to the methylation status of said genes in a reference or control sample. Iso- methylation of the (marker) genes of this invention has the meaning of substantially the same methylation level of said genes, when compared to the methylation status of said genes in a reference or control sample, i.e. the said genes are not differentially methylated compared to the methylation status of said genes in a reference or control sample. In particular embodiments of this invention, the control sample or reference sample is a sample from a healthy prostate. In other particular embodiments of this invention, the control sample or reference sample is a sample from a benign hyperplasia substrate.

In specific embodiments of this invention, the method of this invention comprises PCR analysis of polynucleotide materials of the cells derived from prostatic tissue. In other particular embodiments of this invention, the method of this invention comprises PCR analysis of polynucleotide materials of the cells derived from prostatic fluid.

An embodiment of the present invention is a method of diagnosing a disease state or cell proliferative disorder in the prostate of a subject, said method comprising: (a) analyzing the level of DNA methylation of the regulatory region surrounding the transcription start site (TSS) of certain genes of the panel of genes of this invention (TDRD 1 , RARB, GSTP 1 , APC, CCND2, PTGS2, BCL2, RASSFl, LGALS3, CDH13, PITX2, and HOXD3) or an homologous sequence of said genes in a biological sample isolated from said subject, and (b) comparing said DNA methylation with the DNA methylation in a control sample and/or a benign prostate hyperplasia sample; whereby decreased CpG methylation of the TDRD1 gene and/or CpG methylation or increased CpG methylation in at least one or at least two, three, four, or five genes selected from the group of genes: RARB, GSTP1, APC, CCND2, PTGS2, BCL2, RASSFl, LGALS3, CDH13, PITX2, and HOXD3 relative to the control sample or the benign prostate hyperplasia sample in the regulatory regions surrounding the transcriptional start site of said genes is an indication for prostate cancer and/or an indication of an hormone refractory prostate cancer, androgen-independent prostate cancer (AIPC) or androgen-independent metastatic prostate cancer.

An embodiment of the present invention is a method of diagnosing a disease state or cell proliferative disorder in the prostate of a subject, said method comprising: (a) analyzing the level of DNA methylation of the regulatory region surrounding the transcription start site (TSS) of certain genes of the panel of genes of this invention (TDRD1, RARB, GSTP1, APC, CCND2, PTGS2, BCL2, RASSFl, LGALS3, CDH13, PITX2, and HOXD3) or an homologous sequence of said genes in a biological sample isolated from said subject, and (b) comparing said DNA methylation with the DNA methylation in a control sample and/or a benign prostate hyperplasia sample; whereby decreased CpG methylation of the TDRD1 gene and/or CpG methylation or increased CpG methylation in at least one or at least two, three, four, or five genes selected from the group of genes: RARB, GSTP1, APC, CCND2, PTGS2, BCL2, RASSFl, LGALS3, CDH13, PITX2, and HOXD3 relative to the control sample or the benign prostate hyperplasia sample in the regulatory regions surrounding the transcriptional start site of said genes is an indication of a hormone refractory prostate cancer, an androgen-independent prostate cancer (AIPC) or androgen-independent metastatic prostate cancer.

An embodiment of the present invention is a method of diagnosing a disease state or cell proliferative disorder in the prostate of a subject, said method comprising: (a) analyzing the level of DNA methylation of the regulatory region surrounding the transcription start site (TSS) of certain genes of the panel of genes of this invention (TDRD1, RARB, GSTP1, APC, CCND2, PTGS2, BCL2, RASSFl, LGALS3, CDH13, PITX2, and HOXD3) or an homologous sequence of said genes in a biological sample isolated from said subject, and (b) comparing said DNA methylation with the DNA methylation in a control sample and/or a benign prostate hyperplasia sample; whereby decreased CpG methylation of the TDRD1 gene and/or CpG methylation or increased CpG methylation in at least one or at least two, three, four, or five genes selected from the group of genes: RARB, GSTP1, APC, CCND2, PTGS2, BCL2, RASSFl, LGALS3, CDH13, PITX2, and HOXD3 relative to the control sample or the benign prostate hyperplasia sample in the regulatory regions surrounding the transcriptional start site of said genes is an indication of an androgen-independent metastatic prostate cancer. Further specific embodiments of these previous methods of diagnosis, typing and/or staging can be :

The previous method further comprising a step of analyzing histone (de)acetylation of the gene(s) of this invention in said sample.

The previous method whereby the disease state or cell proliferative disorder is a cancer. The previous method to distinguish between a healthy prostate and a disordered or diseased prostate.

The previous method to distinguish between a benign prostate hyperplasia and a prostate cancer

The previous method to distinguish between an hormone sensitive prostate cancer and an hormone refractory prostate cancer.

The previous method to distinguish between an androgen sensitive prostate cancer or androgen dependent prostate cancer and androgen-independent prostate cancer (AIPC) The previous method to discover an androgen-independent metastatic prostate cancer in a prostate cell or prostate tissue.

The previous method to carry out a prostate cancer grading or prostate cancer staging. The previous method to decide on the proper treatment or proper medicament of the prostate disease state

The previous method to decide on the treatment with a pharmaceutically acceptable DNA methylation inhibitor

The previous method to decide on the treatment with a pharmaceutically acceptable HDAC inhibitor.

The previous method to decide on the treatment to decrease the activity of the EZH2 protein

The previous method to decide on the treatment with a DNA demethylating agent and/or a DNA methyltransferase inhibitor and/or HDAC inhibitor.

The previous method to decide on a prophylactically effective amount of a nutraceutical to treat a subject with a prostate disease status. Numbered statements of the invention are as follows:

1. A method for typing and/or staging and/or predicting outcome of a prostate cell proliferative disorder in a human male subject, the method comprising:

a. analyzing the level of DNA methylation of the regulatory region surrounding the transcription start site (TSS) of the TDRD1 gene in a sample of prostatic tissue or in a biological sample that comprises prostatic cells or prostatic cell components from a human patient; and

b. comparing the methylation level of CpG dinucleotides in the regulatory region surrounding the transcription start sites of TDRD1 of step (a) in the test sample with said level in a reference sample;

wherein the methylation level of CpG dinucleotides in said regulatory region in the TDRDlgene is indicative of the type and/or stage of said prostate cell proliferative disorder.

2. The method according to statement 1, further comprising

a. analyzing the level of DNA methylation of the regulatory region surrounding the transcription start site (TSS) of at least one gene selected from PITX2, RASSFl, and HOXD3 in said test sample; and

b. comparing the methylation level of CpG dinucleotides in the regulatory region surrounding the transcription start sites of the at least one gene of step (a) in the test sample with said level in a reference sample;

wherein the methylation level of CpG dinucleotides in said regulatory region in said at least one gene selected from PITX2, RASSFl, and HOXD3 is further indicative of the type and/or stage of said prostate cell proliferative disorder.

3. The method according to anyone of statements 1 or 2, further comprising

a. analyzing the level of DNA methylation of the regulatory region surrounding the transcription start site (TSS) of at least one gene selected from RARB, GSTP1, APC, CCND2, PTGS2, BCL2, LGALS3, and CDH13 in said test sample; and

b. comparing the methylation level of CpG dinucleotides in the regulatory region surrounding the transcription start sites of the at least one gene of step (a) in the test sample with said level in a reference sample; wherein the methylation level of CpG dinucleotides in said regulatory region in said at least one gene selected from RARB, GSTP1, APC, CCND2, PTGS2, BCL2, LGALS3, and CDH13 is further indicative of the type and/or stage of said prostate cell proliferative disorder.

The method according to anyone of statements 1 to 3, wherein hypomethylation of CpG dinucleotides in the regulatory region surrounding the transcription start sites of TDPvDl is correlated to a prostate cell proliferative disorder of a more aggressive type and/or of a further advanced stage. The method according to statement 2, wherein hypermethylation of CpG dinucleotides in the regulatory region surrounding the transcription start sites of the at least one gene selected from PITX2, RASSF1, and HOXD3 is further correlated to a prostate cell proliferative disorder of a more aggressive type and/or of a further advanced stage. The method according to statement 3, wherein hypermethylation of CpG dinucleotides in the regulatory region surrounding the transcription start sites of the at least one gene selected from RARB, GSTP1, APC, CCND2, PTGS2, BCL2, LGALS3, and CDH13 is further correlated to a prostate cell proliferative disorder of a more aggressive type and/or of a further advanced stage. The method according to anyone of statements 4 to 6 wherein said prostate cell proliferative disorder of a more aggressive type is a biochemical recurrent type of disorder. The method according to anyone of statements 1 to 6 wherein said prostate cell proliferative disorder of a more advanced stage is a high clinical stage disorder of pT stage III or IV. The method according to anyone of statements 2 or 3, wherein iso-methylation of CpG dinucleotides in the regulatory region surrounding the transcription start sites of TDRD1 in combination with iso-methylation or hypomethylation of at least one gene selected from PITX2, RASSF1, HOXD3, RARB, GSTP1, APC, CCND2, PTGS2, BCL2, LGALS3, and CDH13; in particular GSTP1, RARB, PITX2 and HOXD3; is indicative of a methylation-independent type of prostate cell proliferative disorder. The method of statement 1, further comprising

comparing said DNA methylation of TDRDl in said test sample with the DNA methylation in a androgen sensitive prostate cancer sample; and

- wherein decreased methylation of CpG dinucleotides in said regulatory region in the TDRDl gene is indicative of an hormone refractory prostate cancer type, androgen-independent prostate cancer (AIPC) type or androgen-independent metastatic prostate cancer type. The method of statement 10, further comprising:

comparing said DNA methylation of at least one gene selected from PITX2, RASSF1 and HOXD3 and/or at least one gene selected from RARB, GSTP1, APC, CCND2, PTGS2, BCL2, LGALS3, and CDH13 with the DNA methylation in a androgen sensitive prostate cancer sample; and

wherein increased methylation of CpG dinucleotides in said regulatory region of said gene(s) is further indicative of an hormone refractory prostate cancer type, androgen-independent prostate cancer (AIPC) type or androgen-independent metastatic prostate cancer type. The method of any of statements 1 to 11, for use in deciding on the proper treatment or proper medicament dependent on the type and/or stage of said prostate cell proliferative disorder. The method of statement 12, wherein detection of hypomethylation of TDRDl either or not in combination with hypermethylation of at least one gene selected from the list comprising PITX2, RASSF1 and HOXD3, RARB, GSTP1, APC, CCND2, PTGS2, BCL2, LGALS3, and CDH13 is indicative for the decision about the initiation or continuation of a treatment selected from a prostatectomy, a DNA methylation inhibitor or a compound in an effective amount to reduce male hormones. The method of statement 12, wherein detection of iso-methylation of TDRDl either or not in combination with iso-methylation or hypomethylation of at least one gene selected from the list comprising PITX2, RASSF1 and HOXD3, RARB, GSTP1, APC, CCND2, PTGS2, BCL2, LGALS3, and CDH13 is indicative for the decision about the initiation or continuation of a treatment selected from a prostatectomy, histone deacetylase inhibitors, gonadotropin-releasing hormone agonists, neutraceuticals, or radiotherapy. 15. The method of any of statements 1 to 14, wherein methylation is determined using PCR analysis, bisulfite genomic sequencing PCR analysis, Methylation-Specific PCR analysis or an equivalent amplification technique. 16. The method of any of statements 1 to 15, wherein methylation is determined in an assay comprising primers for assessing the presence of methylation in a regulatory region surrounding the TSS of said gene(s). 17. The method of any of statements 1 to 16, wherein at least one primer of the group consisting of methylated specific primers (SEQ. ID NOs 3, 4, 9, 10, 15, 16, 19, 20, 25, 26, 31, 32, 37, 38, 43, 44, 49, 50, 55, 56, 61, 62, 67, 68, 73 and 74) and at least one primer of the group consisting of unmethylated specific primers (SEQ ID NO's 5, 6, 11, 12, 21, 22, 27, 28, 33, 34, 39, 40, 45, 46, 51, 52, 57, 58, 63, 64, 69, 70, 75 and 76) is used. 18. The method of any of statements 1 to 17, wherein the reference sample is a sample from a healthy individual or from an individual having a typical benign hyperplasia prostate. 19. The method of any of statements 1 to 18, wherein said regulatory region surrounding the TTS comprises one or more CpG islands and extends about 1.5 kb upstream to about 1.5 kb downstream from said transcription start site of said gene(s). 20. The method of any of statements 1 to 19, wherein the test and/or reference sample is selected from the list comprising prostatic tissue, prostatic fluid, seminal fluid, ejaculate, blood, urine, prostate secretions, histological slides, and paraffin-embedded tissue. 21. The method of any of statement 1 to 20, further comprising analysing the level of DNA methylation of at least one control plasmid selected from the list comprising SEQ ID N° 91 and 92 either or not in combination with analysing the level of DNA methylation of at least one control plasmid selected from the list comprising SEQ ID N° 77-90 and 93-100.

22. A kit for typing and/or staging a prostate cell proliferative disorder in a human male subject, comprising at least one TDRD1 specific primer selected from the list comprising SEQ ID N°43 and 44 and at least one TDRD1 specific primer selected from the list comprising SEQ ID N° 45 and 46.

23. The kit according to statement 22, further comprising at least one primer selected from the list comprising SEQ ID N° 3, 4, 9, 10, 15, 16, 19, 20, 25, 26, 31, 32, 37, 38, 49, 50,

55, 56, 61, 62, 67, 68 73 and 74 and at least one primer from the list comprising 5, 6, 11, 12, 21, 22, 27, 28, 33, 34, 39, 40, 51, 52, 57, 58, 63, 64, 69, 70, 75 and 76.

24. The kit according to anyone of statement 22 or 23 further comprising at least one control plasmid selected from the list comprising SEQ ID N° 91 and 92 either or not in combination with at least one control plasmid selected from the list comprising N° 77- 90 and 93-100.

Further numbered statements of the invention are as follows: 1. A method of diagnosing a prostate cell proliferative disorder in a human male subject, the method comprising:

a. analyzing the level of DNA methylation of the regulatory region surrounding the transcription start site (TSS) of the TDRD1 gene together with at least one gene selected from PITX2, RASSFl, and HOXD3 in a sample of prostatic tissue or in a biological sample that comprises prostatic cells or prostatic cell components from a human patient; and

b. comparing the methylation level of CpG dinucleotides in the regulatory region surrounding the transcription start sites of the respective genes of step (a) in the test sample with that level in a reference sample;

wherein hypomethylation of CpG dinucleotides in said regulatory region in the

TDRDlgene together with hypermethylation of CpG dinucleotides in said regulatory region of at least one gene selected from PITX2, RASSFl, and HOXD3 indicates the presence of prostate cancer cells or is indicative of a predisposition to, or the incidence of, prostate cancer. The method of statement 1, wherein said at least one gene selected from PITX2, RASSF1, and HOXD3 contains the PITX2, RASSF1 and HOXD3 gene. The method of statement 1 or 2, which further comprises the analysis of the DNA methylation of the regulatory region surrounding the transcription start site (TSS) of at least one gene selected from RARB, GSTP1, APC, CCND2, PTGS2, BCL2, LGALS3, CDH13, and MSMB in a sample of prostatic tissue or in a biological sample that comprises prostatic cells or prostatic cell components from a human patient; and comparing the methylation level of CpG dinucleotides (and non-CpG dinucleotides for MSMB) in the regulatory region surrounding the transcription start sites of at least one gene selected from RARB, GSTP1, APC, CCND2, PTGS2, BCL2, LGALS3, CDH13, and MSMB in the test sample with that level in a reference sample; whereby hypermethylation of CpG dinucleotides (and non-CpG dinucleotides for MSMB) in said region indicates the presence of prostate cancer cells or is indicative of a predisposition to, or the incidence of, prostate cancer. The method of any of statements 1 to 3, wherein the reference sample is from a typical of healthy prostate or from a typical of a benign hyperplasia prostate. The method of any of statements 1 to 4, wherein the region comprises one or more CpG islands and extends about 1.5 kb upstream to about 1.5 kb downstream from the transcription start site of the TDRDl gene and extends about 1.5 kb upstream to about 1.5 kb downstream from the transcription start site for each selected gene. The method of any of statements 1 to 5, wherein the method comprises in step (a) analyzing the level DNA methylation of the genes in a biological sample isolated from said subject, and further comprises in step (b) comparing said DNA methylation with the DNA methylation in a benign prostate hyperplasia sample or a control sample; wherein decreased methylation of CpG dinucleotides in said regulatory region in the TDRDl gene together with increased methylation of CpG dinucleotides in said regulatory region of the selected gene(s) indicates the presence of prostate cancer cells or is indicative of a predisposition to, or the incidence of, prostate cancer. The method of any of statements 1 to 6, wherein the prostate cancer is High-grade. The method of any of statements 1 to 7, wherein the method comprises in step (a) analyzing the level DNA methylation of the genes in a biological sample isolated from said subject, and further comprises in step (b) comparing said DNA methylation with the DNA methylation in a benign prostate hyperplasia sample or a control sample; wherein decreased methylation of CpG dinucleotides in said regulatory region in the TDRD1 gene together with increased methylation of CpG dinucleotides in said regulatory region of the selected gene(s) relative to a control sample is an indication for prostate cancer and (c) comparing said DNA methylation with the DNA methylation in a androgen sensitive prostate cancer sample; and wherein decreased methylation of CpG dinucleotides in said regulatory region in the TDRD1 gene together with increased methylation of CpG dinucleotides in said regulatory region of the selected gene(s) relative to a control sample is an indication of an hormone refractory prostate cancer, androgen-independent prostate cancer (AIPC) or androgen- independent metastatic prostate cancer. The method of any of statements 1 to 8, to discover an androgen-independent metastatic prostate cancer in a prostate cell or prostate tissue.

The method of any of statements 1 to 8, to carry out a prostate cancer grading or prostate cancer staging.

The method of any of statements 1 to 8, to decide on the proper treatment or proper medicament of the prostate disease state.

The method of any of statements 1 to 11, wherein methylation is determined using PCR, or an equivalent amplification technique.

The method of any of statements 1 to 12, wherein methylation is determined by bisulfite genomic sequencing PCR analysis.

The method of any of statements 1 to 13, wherein methylation is determined by Methylation-Specific PCR analysis or an equivalent amplification technique. The method of any of statements 1 to 14, wherein methylation is determined by a diagnostic array, the array comprising primers for assessing the presence of methylation in a regulatory region surrounding the TSS of the genes.

The method of any of statements 1 to 15, wherein at least one primer of the group consisting of methylated specific primers (SEQ. ID NOs 3, 4, 9, 10, 15, 16, 19, 20, 25,

26, 31, 32, 37, 38, 43, 44, 49, 50, 55, 56, 61, 62, 67 and 68) and of the group consisting of unmethylated specific primers (SEQ ID NO's 5, 6, 11, 12, 21, 22, 27, 28, 33, 34, 39, 40, 45, 46, 51, 52, 57, 58, 63, 64, 69 and 70) is used.

The method of any of of statements 1 to 16, wherein PCR analysis is performed on polynucleotide materials of the cells derived from prostatic tissue or prostatic fluid. The method of any of statements 1 to 17, wherein PCR analysis is performed on polynucleotide materials of the cells derived from seminal fluid or from ejaculate. The method of any of statements 1 to 17, wherein PCR analysis is performed on polynucleotide materials of the cells derived from body fluids such as blood, urine, ejaculates or prostate secretions.

The method of any of statements 1 to 17, wherein PCR analysis is performed on polynucleotide materials of the cells derived from prostate tissue from histological slides or biopsies or paraffin-embedded tissue.

The method of any of statements 1 to 17 or statement 20, wherein PCR analysis is performed on polynucleotide materials of the cells of prostatic tissue from biopsy or from surgical resection.

The method of any of statements 1 to 21, for distinguishing between an androgen- independent or androgen-refractory prostate cancer and/or an androgen-sensitive prostate cancer in a tissue of a subject encountered by prostate cancer according to statement 1, wherein differential methylation is observed when compared to the methylation status of the regulatory region surrounding the transcription start site of the genes from a androgen-sensitive prostate cancer cell or a normal cell, which differential methylation is hypomethylation in the case of the TDRDl gene and hypermethylation in the other selected gene(s) and is indicative for androgen- refractory prostate cancer.

The method of any of statements 1 to 22, wherein detection of hypomethylation in the case of the TDRDl gene and hypermethylation in the other selected gene(s) is indicative for the grade and/or stage of the prostate proliferative disorder. 24. The method of any of statements 1 to 23, wherein detection of hypermethylation is indicative for the decision about a treatment with a DNA methylation inhibitor.

25. The method of any of statements 1 to 24, wherein detection of hypermethylation is indicative for the decision about the initiation or continuation of treating with a compound in an effective amount to reduce male hormones.

26. The method of any of statements 1 to 25, further comprising analysing aceteylation of the gene histones for distinguishing between an androgen-independent or androgen- refractory prostate cancer and/or an androgen-sensitive prostate cancer.

27. A method of treating prostate cancer in a human patient, comprising administration of a DNA demethylating agent, wherein the human patient has been selected for treatment on the basis of a method as claimed in any preceding statement.

28. A kit for detecting a predisposition to, or the incidence of, prostate cancer in a sample comprising at least one primer pair for determining the methylation status of TDRD1 and at least one primer pair for determining the methylation status of at least one gene selected from RARB, GSTP1, APC, CCND2, PTGS2, BCL2, RASSF1, LGALS3, CDH13, PITX2, HOXD3, and MSMB.

29. The kit of statement 28, which comprises means for detecting differential methylation in said genes, wherein hypomethylation of CpG dinucleotides in the regulatory region surrounding the transcription start site (TSS) of the TDRD1 gene together with hypermethylation of CpG dinucleotides (and non-CpG islands for MSMB) in the regulatory region surrounding the transcription start site (TSS) of at least one gene selected from RARB, GSTP1, APC, CCND2, PTGS2, BCL2, RASSF1, LGALS3, CDH13, PITX2, HOXD3, and MSMB is indicative of a predisposition to, or the incidence of, prostate cancer.

30. The kit of statement 28 or 29, wherein the said at least one gene is selected from PITX2, RASSF1 and HOXD3.

Further scope of applicability of the present invention will become apparent from the detailed description given hereinafter. However, it should be understood that the detailed description and specific examples, while indicating preferred embodiments of the invention, are given by way of illustration only, since various changes and modifications within the spirit and scope of the invention will become apparent to those skilled in the art from this detailed description. It is to be understood that both the foregoing general description and the following detailed description are exemplary and explanatory only and are not restrictive of the invention, as claimed.

DESCRIPTION OF THE FIGURES

Figure 1 : Figure 1A is a schematic representation of the strategy we have chosen for the analysis of methylation status of different potential prostate cancer biomarkers. For the fast estimation of promoter methylation a melting curve assay was applied to a set of model genotypes corresponding to non-cancerous DNA and tissues (human genomic DNA (HG DNA), cell lines PZ-HPV7, BPHl) and PCa cell lines (androgene-sensitive LNCaP and androgene-insensitive PC-3 and DU 145). Melting curve assay implies amplification of a part of the gene promoter with methylation independent primers after bisulphite conversion of DNA, followed by registration of melting profile of the resulting amplicons. Differences in melting profile help to discriminate between initially methylated and nonmethylated DNA templates. Results of the analysis of the APC gene are shown in Fig. IB.

Figure 2: Results of the bisulphite sequencing of the PCR- fragments covering MAGE Ά 2 CpG- island around the transcription site, obtained from whole blood human genomic DNA (HG DNA) and LNCaP cell line. Numbers from -2 to 5 and from 12 to 16 represent positions of CpG-dinucleotides relative to the transcription start site (TSS). The selected CpG-island is completely hypomethylated in cancer LNCaP cell line in comparison with whole blood human genomic DNA. Figure 3: Graphs showing PCR cycles (X-axis) plotted against the fluorescence intensity of the PCR product accumulated in EvaGreen® reaction mixture (Y-axis) using 100% M and 100% U plasmid standards as a template tested with M and U primers for APC. The M and U reactions were 100% specific since M primers did not cross-react with U standard and vice versa. No primer dimers were observed in "no template" control. Specificity of primers is confirmed by melting curve analysis: melting temperature of the PCR product generated from 100% M template is shifted to the right and corresponds to that of the completely methylated sequence, as indicated by the melting peak; melting temperature of the PCR product generated from 100% U template is shifted to the left and corresponds to that of the completely unmethylated sequence, as indicated by the melting peak. Figure 4: Graphs showing validation of primers for qMSP and qUSP using APC methylated (M) and unmethylated (U) primer sets to amplify serially diluted plasmid standards in EvaGreen® PCR mixture. Efficency 0.95 - 1.00 reflects a 2-fold amplification of DNA per cycle. The correlation coefficient (R^A2) shows linearity (0.998) over the range of DNA concentrations.

Figure 5: Procedure of validation of primer sets for the two-step Quantitative multiplex nested-MSP assay.

Figure 6: Schematic representation of the protocol for quantitative multiplex nested-MSP analysis. In Multiplex PCR step a mixture of gene-specific methylation-independent primer pairs is used to co-amplify 80-180 bp fragments of CpG islands covering regulatory elements of the selected genes. In the quantitative step two real time PCRs (qMSP and qUSP) are performed for each gene separately with primer sets specific for methylated (M) and unmethylated (U) sequences using the DNA template derived from Multiplex PCR step (diluted 1 :500 in sterile water). Serial dilutions 3x 10² - 3x 10⁶ of plasmid standards with cloned gene fragments corresponding to completely methylated (M) and completely unmethylated (U) sequences are used to generate separate standard curves and quantify M and U fragments. The percent methylation for each gene in the panel is calculated as %M = [M / (U + M)] 100.

Figure 7: Gray-scale representation of the levels of 16 genes promoter hypermethylation as determined by the invention methods employing quantitative multiplex nested-MSP on prostate cell lines, prostate tissues and HG DNA. Intensity of color correlates with the degree of methylation, also indicated by number (%). For the TDRD1 and MAGEA2 genes reverse methylation value is presented (100 - % of methylation).

Figure 8: Gray-scale representation of the levels of 16 genes promoter hypermethylation as determined by the invention methods employing quantitative multiplex nested-MSP on matched tumor/adjacent benign prostate tissue samples from 7 patients. Intensity of color correlates with the degree of methylation, also indicated by number (%). For the TDRD1 and MAGEA2 genes the reverse methylation values are presented (100 - % of methylation).

Figure 9: Average tumor volume measured (A) in ml and (B) in % of the total prostate gland volume in patients with low (LM) and high (HM) methylation of tumor DNA. Low and high methylation levels are discriminated based on the median methylation value for each gene. Study group: PCa2 cohort (N=63). Genes: 1 - RARB, 2 - GSTP1, 3 - CCND2, 4 - PTGS2, 5 - APC, 6 - LGALS3, 7 - TDRDl, 8 - RASSF1, 9 - PITX2, 10 - HOXD3, 11 - CDH13. The differences are significant at * P < 0.05, ** P < 0.01 and *** P < 0.001. Figure 10: Results of Kaplan-Meier analysis for biochemical progression-free survival probability within 16 years after radical prostatectomy in groups of patients with high (HM - above the cutoff methylation value) and low (LM - below the cutoff methylation value, indicated on the graph) degree of HOXD3 (A) and TDRDl (B) methylation in PCa tumors from the PCa2 cohort. Figure 11 : Results of Kaplan-Meier analysis for biochemical progression- free survival probability within 16 years after radical prostatectomy in groups of patients with high (HM - above the cutoff methylation value) and low (LM - below the cutoff methylation value, indicated on the graph) degree of PITX2 (A) and RASSF1 (B) methylation in PCa tumors from the PCa2 cohort. Figure 12: Results of Kaplan-Meier analysis for clinical failure (CF) probability within 16 years after radical prostatectomy in groups of patients with high (HM - above the cutoff methylation value) and low (LM - below the cutoff methylation value, indicated on the graph) degree of PITX2 methylation in PCa tumors from the PCa2 cohort.

DETAILED DESCRIPTION OF THE INVENTION As used herein, the singular forms "a", "an" and "the" include plural referents unless the context clearly dictates otherwise.

"Disease state" as used herein means any disease, disorder, condition, symptom, or indication.

As used herein, the term "cell proliferative disorder" refers to conditions in which the unregulated and/or abnormal growth of cells can lead to the development of an unwanted condition or disease, which can be cancerous or non-cancerous. The detection of the cell proliferative disorder may be by way of routine examination, screening for a cell proliferative disorder or pre-stadia such cell proliferative disorder, monitoring and/or staging the state and/or progression of the cell proliferative disorder, assessing for recurrence following treatment, and monitoring the success of a treatment regimen. In a preferred embodiment, the cell proliferation disorder is cancer.

As used herein, the term "cancer" concerns malignant neoplasm, malignant tumor or invasive tumor and also can include solid neoplasm or solid tumors cancers. Cancers are classified by the type of cell that resembles the tumor and, therefore, the tissue presumed to be the origin of the tumor. Examples of general categories include: Carcinoma: Malignant tumors derived from epithelial cells. This group represents the most common cancers, including the common forms of breast, prostate, lung and colon cancer. Sarcoma: Malignant tumors derived from connective tissue, or mesenchymal cells. Lymphoma and leukemia: Malignancies derived from hematopoietic (blood-forming) cells Germ cell tumor: Tumors derived from totipotent cells. In adults most often found in the testicle and ovary; in fetuses, babies, and young children most often found on the body midline, particularly at the tip of the tailbone; in horses most often found at the poll (base of the skull). Blastic tumor: A tumor (usually malignant) which resembles an immature or embryonic tissue. In a preferred embodiment, the cancer is prostate cancer.

"Hormone refractory prostate cancer" and in particular "androgen-independent prostate cancer (AIPC)" has to be understood for the meaning of this invention as prostate cancer that has become refractory, that is, it no longer responds to hormone therapy.

"Prostate cancer grading" or "typing" as used herein means describing how abnormal or aggressive the cancer cells appear. The grade/type helps to predict long-term results, response to treatment and survival. In the art there is for instance the Gleason scale that is the most common scale used for grading prostate cancer. This system assigns cancer cells a score from 1 to 10, by combining the two most common patterns of cells to give a total score (i.e., 3 + 4 = grade 7). Scores generally range between 4 and, most commonly, 6 or 7. These scores are broken down into three main levels: Low-grade (well differentiated): This type of slow- growing cancer has an appearance most like normal prostate cells and is the least dangerous. It has a Gleason score of 4 or less. Intermediate grade (moderately differentiated): This type is somewhere between the low- and high-grade cancers and the most common of the three. Depending on PSA level and tumor volume, it can act like a high- or low-grade cancer. It has Gleason score between 4 and 7. High-grade (poorly differentiated): This type of cancer has an appearance least like normal prostate cells. It is the most deadly since it is very aggressive and grows very fast - even into surrounding areas - such as lymph nodes and bones. These cancer cells also tend to be large, hard to treat, and reappear more frequently. They have a Gleason score between 8 and 10 (Antoinette S Perry et al Endocrine-Related Cancer 13 (2) 357-377) .

"Prostate cancer staging " or "staging" as used herein concerns how much and where the cancer is located. The more cancer there is in the body, the more likely it is to spread and less likely that treatments will work. Therefore, the more advanced stages can affect long-term results and survival. According an older prostate cancer staging the prostate cancer is broken down into four primary stages for instance the four ABCD stages of staging to gauge the severity of prostate cancer to describe the detection and location of the cancer. Stage A: Cancer found when not suspected or due to a high PSA level, Stage B: Cancer found due to abnormal digital rectal exam and is held in the prostate, Stage C: Cancer that has spread to the tissues outside of the prostate, Stage D: Cancer that has spread to the lymph nodes or bone. A particular system in the art which replaced the ABCD staging system of prostate cancer to give an even more accurate description of the cancer is the TNM grading system. "T" describes the tumor and uses different numbers to explain how large it is; "N" stands for nodes and tells whether the cancer has spread to the lymph nodes; "M" means metastatic, and tells whether the cancer has spread throughout the body. There are various T Status stages : Stage Tl : Microscopic tumor confined to prostate and undetectable by a digital rectal exam (DRE) or ultrasound; Stage Tla: Tumor found in 5% or less of prostate tissue sample; Stage Tib: Tumor found in more than 5% of a prostate tissue sample; Stage Tic: Tumor is identified by needle biopsy as a follow-up to screening that detected elevated PSA results; Stage T2: Tumor confined to prostate and can be detected by DRE or ultrasound; Stage T2a: Tumor involves less than half of one lobe of the prostate, and can usually be discovered during DRE exam; Stage T2b: Tumor involves more than half of one lobe of the prostate, and can usually be felt during DRE exam; Stage T2c: Tumor involves both lobes of the prostate and is felt during a DRE exam; Stage T3 : Tumor has spread to surrounding tissues or to the seminal vesicles; Stage T3a: Tumor has spread to outside of the prostate on only one side; Stage T3b: Tumor has spread to outside of the prostate on both sides; Stage T3c: Tumor has spread to one or both of the seminal tubes; Stage T4: Tumor is still within the pelvic region but may have spread to organs near the prostate, such as the bladder; Stage T4a: Tumor has spread beyond the prostate to any or all of the bladder neck, the external sphincter, and/or the rectum and Stage T4b: Tumor has spread beyond the prostate and may affect the levator muscles (the muscles that help to raise and lower the organ) and/or the tumor may be attached to the pelvic wall and various N Status stages : Stage NO: Cancer cells have spread, but not yet to pelvic lymph nodes; Stage Nl : Cancer cells have spread to a single lymph node in the pelvic area and are 2 cm (approximately 3/4 of one inch) or less in size; Stage N2: Cancer cells have spread either to a single lymph node and are more than 2 cm but less than 5 cm (approximately 2 inches) in size, or the prostate cancer cells are found in more than one lymph node and are no larger than 5 cm in size; Stage N3: Cancer cells have spread to the lymph nodes and are larger than 5 cm in size and various M Status stages: Stage MO: Cancer cells have spread, but only regionally in the pelvic area & Stage Ml : Cancer cells have spread beyond the pelvic area to other parts of the body (Dr F. H. Schroder et al. The Prostate Volume 21, Issue S4 , Pages 129 - 138, 20 Jul 2006). As used herein, the term "effective amount" refers to an amount of a compound, or a combination of compounds, of the present invention effective when administered alone or in combination as an anti-proliferative agent. For example, an effective amount refers to an amount of the compound present in a formulation or on a medical device given to a recipient patient or subject sufficient to elicit biological activity, for example, anti-proliferative activity, such as e.g., anti-cancer activity or anti-neoplastic activity. The combination of compounds optionally is a synergistic combination. Synergy, as described, for example, by Chou and Talalay, Adv. Enzyme Regul. vol. 22, pp. 27-55 (1984), occurs when the effect of the compounds when administered in combination is greater than the additive effect of the compounds when administered alone as a single agent. In general, a synergistic effect is most clearly demonstrated at sub-optimal concentrations of the compounds. Synergy can be in terms of lower cytotoxicity, or increased anti-proliferative effect, or some other beneficial effect of the combination compared with the individual components.

"A therapeutically effective amount" as used herein means the amount of a compound that, when administered to a mammal for treating a disease, is sufficient to effect such treatment for the disease. The "therapeutically effective amount" will vary depending on the compound, the disease and its severity and the age, weight, etc., of the mammal to be treated. A therapeutically effective amount of one or more of the compounds can be formulated with a pharmaceutically acceptable carrier for administration to a human or an animal. Accordingly, the compounds or the formulations can be administered, for example, via oral, parenteral, or topical routes, to provide an effective amount of the compound. In alternative embodiments, the compounds prepared in accordance with the present invention can be used to coat or impregnate a medical device. The term "prophylactically effective amount" as used herein means an effective amount of a compound or compounds, of the present invention that is administered to prevent or reduce the risk of unwanted cellular proliferation.

"Pharmacological effect" as used herein encompasses effects produced in the subject that achieve the intended purpose of a therapy. In one preferred embodiment, a pharmacological effect means that primary indications of the subject being treated are prevented, alleviated, or reduced. For example, a pharmacological effect would be one that results in the prevention, alleviation or reduction of primary indications in a treated subject. In another preferred embodiment, a pharmacological effect means that disorders or symptoms of the primary indications of the subject being treated are prevented, alleviated, or reduced. For example, a pharmacological effect would be one that results in the prevention or reduction of primary indications in a treated subject.

"Prostate biopsy" as used herein is a procedure in which small samples are removed from a man's prostate gland to be tested for the presence of cancer. It is typically performed when the scores from a PSA blood test rise to a level that is associated with the possible presence of prostate cancer.

A subject from which a biological sample can be obtained for analysis according to the invention is an animal such as a mammal, e.g. a dog, cat, horse, cow, pig, sheep, goat, primate, rat, or mouse. A preferred subject is a human being, particularly a patient suspected of having or at risk for developing a cell proliferative disorder such as a prostate cancer, or a patient with such a cell proliferative disorder such as a prostate cancer.

"Treating", includes any effect, e.g., lessening, reducing, modulating, or eliminating, that results in the improvement of the condition, disease, disorder, etc. "Treating" or "treatment" of a disease state includes: (1) preventing the disease state, i.e. causing the clinical symptoms of the disease state not to develop in a subject that may be exposed to or predisposed to the disease state, but does not yet experience or display symptoms of the disease state; (2) inhibiting the disease state, i.e., arresting the development of the disease state or its clinical symptoms; or (3) relieving the disease state, i.e., causing temporary or permanent regression of the disease state or its clinical symptoms. "Prostatic cell component(s)" mean any part of or component of a prostatic cell or cell compartment derived from a prostatic cell, including exosomes and preferably comprising DNA of said prostatic cell.

By "homologous sequence" is meant, a nucleotide sequence that is shared by one or more polynucleotide sequences, such as genes, gene transcripts and/or non-coding polynucleotides. For example, a homologous sequence can be a nucleotide sequence that is shared by two or more genes encoding related but different proteins, such as different members of a gene family, different protein epitopes, different protein isoforms or completely divergent genes, such as a cytokine and its corresponding receptors. A homologous sequence can be a nucleotide sequence that is shared by two or more non-coding polynucleotides, such as noncoding DNA or RNA, regulatory sequences, introns, and sites of transcriptional control or regulation. Homologous sequences can also include conserved sequence regions shared by more than one polynucleotide sequence. Homology does not need to be perfect homology (e.g., 100%), as partially homologous sequences are also contemplated by the instant invention (e.g., 99%, 98%, 97%, 96%, 95%, 94%, 93%, 92%, 91%, 90%, 89%, 88%, 87%, 86%, 85%, 84%, 83%, 82%, 81%, 80% etc.).

With the "regulatory region surrounding the transcription start site (TSS)" is meant a regulatory region located upstream or 5' to the TSS and/or a regulatory region around the TSS and/or a regulatory region located downstream or 3' to the TSS of the concerned gene. The location of the concerned region can vary from 15Kbp upstream to 15 Kbp downstream of the TSS. Thus the region under investigation may correspond to all or part of the promoter region of the concerned gene. Alternatively, the region under investigation corresponds an exon and/or intron region and/or TSS region of the concerned gene. The region of the concerned gene is preferably between about 1500 bp upstream and about 1500 bp downstream from the TSS of the gene. In a particular embodiment of this invention, said region extends from -1500 bp and +1500 bp from the TSS of the gene.

The term "promoter" refers to the regulatory region located upstream, or 5' to the structural gene and/or TSS. Such a region extends typically between approximately 5 Kb, 500 bp or 150 to 300 bp upstream from the transcription start site of the concerned gene. For all the genes of this invention, we identified at least 1 CpG islands (genomic regions that contain a high frequency of CG dinucleotides) surrounding the transcriptional start site. By "conserved sequence region" is meant, a nucleotide sequence of one or more regions in a polynucleotide does not vary significantly between generations or from one biological system, subject, or organism to another biological system, subject, or organism. The polynucleotide can include both coding and non-coding DNA and RNA. By "control plasmid" is meant, a plasmid comprising a sequence insert, corresponding to methylated or unmethylated DNA sequence after bisulphite conversion, which is representative for the respective methylated or unmethylated state of its corresponding gene. Said insert is in particular selected from the list comprising SEQ ID N° 77-100, as depicted in table 13. Serial dilutions of plasmid standards with inserts corresponding to completely methylated (M) and completely unmethylated (U) sequences can be used to generate separate standard curves and quantify M and U fragments. The percent methylation for each gene in the panel can then be calculated as %M = [M / (U + M)] x 100.

By "EZH2" as used herein is meant, the EZH2 gene and any polycomb group protein EZH2 protein, peptide, or polypeptide having any polycomb group protein EZH2 activity, such as encoded by EZH2 or any other polycomb group protein EZH2 transcript derived from an EZH2 gene. The term EZH2 also refers to nucleic acid sequences encoding any polycomb group protein EZH2 protein, peptide, or polypeptide having EZH2 activity. The term "EZH2" is also meant to include other EZH2 encoding sequence, such as other EZH2 isoforms, mutant EZH2 genes, splice variants of EZH2 genes, and EZH2 gene polymorphisms. The polycomb group protein enhancer of zeste homolog 2 (EZH2) is overexpressed in hormone-refractory, metastatic prostate cancer (Varambally et al, 2002 , Nature, 419, 624-629). An example of such EZH2 is for instance the unprocessed precursor with entry in the UniProtKB/Swiss-Prot and with primary accession number Q 15910, Protein name Enhancer of zeste homolog 2 Synonym ENX-1 Gene name Name: EZH2 and the sequence of the unprocessed precursor (Length: 746 AA (This is the length of the unprocessed precursor) Molecular weight: 85363 Da .

The genes of this invention comprise all the genes, variants, alternative names of said genes that are known to a person skilled in the art. Some of the alternative names of the genes of this invention are summarized here below. Description of the genes of this invention:

The gene MSMB encodes, microseminoprotein, beta (also known as MSP; PSP; IGBF; MSPB; PN44; PRPS; PSP57; PSP94; PSP-94) which is a member of the immunoglobulin binding factor family. PSP94 encoded by the MSMB gene is a tumor suppressor. The Prostate secretory protein of 94 amino acids (PSP94), encoded by the highly prostate-specific MSMB gene, is one of the three major proteins secreted in the seminal fluid, together with PSA and Prostatic Acid Phosphatase (PAP). It has been shown that PSP94 decreases tumor growth in a syngenic in vivo model of PrCa (Shukeir et al. (2003) Cancer Res. 63:2072-8) and suppresses PC-3 cell clonogenic growth as well as the growth of PC-3 xenografts (Garde SV, et al. (1999) Prostate 38: 118-25). Interestingly, the peptide PCK3145, derived from PSP94 and patented by Ambrilia (Biopharmaceutical company, Quebec, Canada), is currently being clinically tested for the treatment of advanced PrCa (Hawkins RE, Daigneault L, Cowan R, Griffiths R, Panchal C, et al. (2005) Clin Prostate Cancer. 4:91-9) The MSMB gene is approximately 13 kb in length, comprises 4 exons and 3 introns, and encodes a transcript of 572 nucleotides. It is synthesized by the epithelial cells of the prostate gland and secreted into the seminal plasma. This protein has inhibin-like activity. It may have a role as an autocrine paracrine factor. The expression of the encoded protein is found to be decreased in prostate cancer. Two alternatively spliced transcript variants encoding different iso forms are described for this gene. One transcript variant (Homo sapiens microseminoprotein, beta- (MSMB), transcript variant PSP94) has been deposited in NCBI under the accession number NM 002443.2, and the other variant (Homo sapiens microseminoprotein, beta- (MSMB), transcript variant PSP57, mRNA) has been deposited in CBI under the accession number NM l 38634.1. Accordingly, in a first aspect the invention provides a method for typing, staging, predicting outcome and/or identifying a prostate cell proliferative disorder in a human male subject, the method comprises:

- providing a sample of prostatic tissue and/or biological fluid of the prostate from a human patient susceptible to a prostate cancer and,

- analyzing the sample for the presence of methylation (on CpG dinucleotides) in a regulatory region surrounding the transcription start site or promoter region of certain genes of the panel of genes of this invention (TDRD1, RARB, GSTP1, APC, CCND2, PTGS2, BCL2, RASSF1, LGALS3, CDH13, PITX2, and HOXD3),

wherein the presence of hypomethylation in said region of the TDRDl gene and/or hypermethylation in this region in at least one or at least two, three, four, or five genes selected from the group of genes: RARB, GSTP1, APC, CCND2, PTGS2, BCL2, RASSF1, LGALS3, CDH13, PITX2, and HOXD3 relative to the methylation status of the corresponding region in the control sample or the benign prostate hyperplasia sample is indicative of prostate cancer.

Preferrably, the method comprise the steps of:

a) obtaining a biological sample from the subject; b) determining the methylation state of CpG island upstream and/or downstream of the TSS region and/or in the promoter region of the panel of genes of this invention (TDRDl, RARB, GSTP1, APC, CCND2, PTGS2, BCL2, RASSF1, LGALS3, CDH13, PITX2, and HOXD3); and

c) identifying hypermethylation of the region(s), wherein hypermethylation (on CpG dinucleotides) is identified as being different when compared to the same region(s) of the gene or associated regulatory region in a subject not having the prostate cellular proliferative disorder,

wherein detection of hypomethylation of the TDRDl gene and/or hypermethylation in at least one or at least two, three, four, or five genes selected from the group of genes: RARB, GSTP1, APC, CCND2, PTGS2, BCL2, RASSF1, LGALS3, CDH13, PITX2, and HOXD3 is indicative of a predisposition to, or the incidence of, prostate cancer.

In a particular embodiment this invention relates to a diagnosis for a prostate proliferation disorder specifically by additionally identification hypermethylations of non CpG dinucleotides (for instance hypermethylation of CpA, CpT or CpC dinucleotides) in particular in a regulatory region surrounding the transcriptional start site of the beta- microseminoprotein (MSMB) gene or in particular upstream of the promoter region or in the promoter region of the beta-microseminoprotein (MSMB) gene. Such diagnosis on cell or tissue samples of prostate allows specifically to distinguish between tissues benign prostate hyperplasia and prostate cancer.

More specifically this invention also relates to a diagnosis for a prostate proliferation disorder specifically by identification hypermethylations in genomic regions that contain a high frequency of CG dinucleotides (CpG islands) and in particular in a regulatory region surrounding the transcriptional start site of at least one, two, three, four or five of the genes selected from the list: RARB, GSTP1, APC, CCND2, PTGS2, BCL2, RASSFl, LGALS3, CDH13, PITX2, HOXD3 and MSMB, wherein for MSMB hypermethylation is in non-CpG islands. Hypermethylations in genomic regions that contain a high frequency of CG dinucleotides (CpG islands) and in particular upstream of the promoter region or in the promoter region of said gene(s) are indicative of a prostate - proliferative disorder that allows to distinguish between a hormone sensitive and a hormone refractory prostate cell proliferative disorder. A more particular aspect of the present invention relates to a diagnostic indicator of an androgen hormone refractory prostatic tissue cellular proliferative disorder, for instance an androgen hormone refractory prostate cancer (PrCa). An other aspect of the present invention relates to a diagnostic indicator of 1) a benign prostate hyperplasia or a prostate cancer and 2) in case of a prostate cancer of an androgen hormone sensitive prostatic tissue cancer or an androgen hormone refractory prostatic tissue cancer. The sample for use in such methods can be any suitable sample such as prostatic tissue, prostatic fluid, seminal fluid, ejaculate, blood, urine, prostate secretions, histological slides, and paraffin-embedded tissue, and is preferably a tissue sample. Prostate biopsy is a procedure in which small samples are removed from a man's prostate gland to be tested for the presence of cancer. It is typically performed when the scores from a PSA blood test rise to a level that is associated with the possible presence of prostate cancer. A biopsy thus provides a specific example of a biological sample for use in present methods. Examination of the condition of the prostate may be performed transrectally, through the ureter or through the perineum. The most common procedure is transrectal, and may be done with tactile finger guidance,( Ghei, M; Pericleous S et al (2005 Sep). Ann R Coll Surg Engl 87 (5): 386-7.) or with ultrasound guidance. If cancer is suspected, a biopsy is offered. During a biopsy tissue samples from the prostate are obtained for instance via the rectum. A biopsy gun can be used to insert and remove special hollow-core needles (usually three to six on each side of the prostate) in less than a second. Suitable samples for diagnostic, prognostic, or personalised medicinal uses can be obtained from surgical samples, such as biopsies or surgical resection. However, other suitable samples for use in the methods of present invention comprise fine needle aspirates, paraffin embedded tissues, frozen tumor tissue samples, fresh tumor tissue samples, fresh or frozen body fluid. Examples of body fluids include prostatic fluids, blood samples, serum, plasma, urine, ejaculate, wash or lavage fluid. In fact, any tissue or fluid containing cells or nucleic acid, preferably DNA, derived from cells of the prostate is a suitable reagent for use in the methods of present invention. Present methods preferably also include the step of obtaining the suitable sample. Cells may need to be lysed for release of the nucleic acid. The nucleic acid may need to be cleared of proteins or other contaminants, e.g. by treatment with enzymes. The nucleic acid may also need to be concentrated prior to further use in the method of the invention, in particular when the nucleic acid is derived from bodily fluids.

As shown by present invention, the above mentioned methods for identifying prostate tumor cells also allow distinguishing hormone sensitive from hormone refractory prostate cancers. Thus in a particular aspect, the present invention provides for an in vitro method for distinguishing a hormone independent proliferative disorder or hormone refractory proliferative disorder from a hormone sensitive proliferative disorder in tissue and/or in at least one cell obtainable from tissue of the prostate from a subject. Such prognostic/diagnostic method comprises contacting a DNA of a tissue or a DNA of a biological fluid with a reagent which detects the methylation status of the promoter region of at least one, two, three, four or five genes of the panel of genes of this invention (TDRD1, RARB, GSTP1, APC, CCND2, PTGS2, BCL2, RASSF1, LGALS3, CDH13, PITX2, and HOXD3), wherein hypomethylation of TDRD1 and/or hypermethylation of at least one, two, three, four or five genes selected from the group: RARB, GSTP1, APC, CCND2, PTGS2, BCL2, RASSF1, LGALS3, CDH13, PITX2, and HOXD3, as compared to the methylation status of the promoter region or upstream of the promoter region of said gene(s) from said group from a normal cell or compared to the methylation status of promoter region or upstream of the promoter region of said gene(s) from said group from cells or of tissue of a prostate with steroidal hormone sensitive proliferative disorder, is indicative of said steroidal hormone refractory proliferative disorder.

The test is particularly suitable to distinguish between hormone refractory and homone sensitive and in particular for androgen sensitive and androgen-refractory prostate proliferative disorders and to distinguish between benign prostate hyperplasia and prostate cancer.

In one embodiment, the invention provides a method for distinguishing between androgen sensitive and androgen-refractory prostate cancer by contacting a cellular component of a prostate tissue sample or another sample with a reagent which detects the methylation status of at least one, two, three, four or five genes of the panel of genes of this invention (TDRD1, RARB, GSTP1, APC, CCND2, PTGS2, BCL2, RASSF1, LGALS3, CDH13, PITX2, and HOXD3) in the promoter or upstream of the promoter region of said gene(s). As aforementioned, methylation sensitive restriction endonuclease can be utilized to identify a hypermethylated promoter or upstream region of the genes of this invention, for example.

Other approaches for detecting methylated CpG dinucleotide motifs use chemical reagents. In particular chemical reagents that selectively modify the methylated or non-methylated form of CpG dinucleotide motifs can be used in the methods of present invention. Such chemical reagents include bisulphite ions. Sodium bisulphite converts unmethylated cytosine to uracil but methylated cytosines remain unconverted. Analysis of the nucleic acid sequence after bisulfite conversion indicates if the original nucleic acid was all or not methylated.

Multiple techniques for analysing the methylation status of CpG dinucleotide motifs in CpG islands are known in the art. They comprise without limitation sequencing, methylation- specific PCR (MS-PCR), McMS-PCR, MLPA, QAMA, MSRE-PCR, MethyLight, HeavyMethyl, ConLight-MSP, BS-MSP, COBRA, McCOBRA, MS-SNuPE, MS-SSCA, PyroMethA, MALDI-TOF, MassARRAY, ERMA, QBSUPT, MethylQuant, Quantitative PCR sequencing, oligonucleotide-based microarray systems, Pyrosequencing, and Meth- DOP-PCR. A review of techniques for the detection of the methylation state of a gene is given for instance in Oral Oncology, 2006, Vol. 42, 5-13 and references cited therein.

A preferred technique for the detection and/or quantification of methylated DNA is the Methylation Specific PCR (MSP) technique. This technique can be used in end-point format, wherein the presence of methylated DNA is for instance detected by electroforesis or by the use of dyes such as SYBR Green I or Ethidium Bromide that bind double-stranded DNA that accumulates during the amplification reaction. Alternatively, the method is based on the continuous optical monitoring of an amplification process and utilises fluorescently labeled reagents. Their incorporation in a product can be quantified as the reaction processes and is used to calculate the copy number of that gene or sequence region in the sample. The quantification of the amplification product may require the use of controls to avoid false negativity/positivity of the reaction. Particularly suitable for the quantification of the amplification product are reference genes (e.g. beta-actin) whose methylation status is known, and/or DNA standards (e.g. methylated or unmethylated standards).

Accumulation of an amplification product can be monitored through the incorporation of labeled reagents. Some techniques use labeled primers; others rely upon the use of labeled probes to monitor the amplification product. Real-time quantitative methylation specific PCR techniques comprise the use of Amplifluor primers and/or Molecular Beacon probes and/or Fret probes and/or Scorpion primers and/or Taqman probes and/or oligonucleotide blockers (eg. HeavyMethyl approach) and/or DzyNA primers. All these probes and primers have been described and their mode of action is well known in the art. In a preferred embodiment, the methods of the invention use unmethylated specific primers indicated by SEQ ID NO's 5, 6, 11, 12, 21, 22, 27, 28, 33, 34, 39, 40, 45, 46, 51, 52, 57, 58, 63, 64, 69 and 70 and/or methylated specific primers indicated by SEQ ID NO's 3, 4, 9, 10, 15, 16, 19, 20, 25, 26, 31, 32, 37, 38, 43, 44, 49, 50, 55, 56, 61, 62, 67 and 68.

Alternatively to PCR, other amplification methods such as NASBA, 3SR, TMA, LCR, selective amplification of target polynucleotide sequences (US Patent No. 6,410,276), consensus sequence primed polymerase chain reaction (US Patent No 4,437,975), arbitrarily primed polymerase chain reaction (WO 90/06995), invader technology, strand displacement technlology, and nick displacement amplification (WO 2004/067726) may be used to amplify the appropriate nucleic acid.

In a further preferred embodiment, bisulphite sequencing is utilised in order to determine the methylation status of the MSMB gene. Primers may be designed in both the sense and antisense orientation to direct sequencing across the relevant region of the genes of this invention. Said primers can easily be designed by a person skilled in the art.

These amplification primers, amplification probes and sequencing primers form a further aspect of the invention.

This invention provides prognostic and/or diagnostic tools or means to determine a prostate cancer and to distinguish between androgen sensitivity and androgen independency of such prostate cancer. Methylation changes are not only ideal for screening purposes, but also interesting targets for monitoring staging or grading of the cancer. Methods for identifying a prostate cell proliferative disorder in a subject, can comprise the steps of: a) obtaining a biological sample from the subject; b) determining the methylation state of CpG island upstream and/or downstream of the TSS region and/or in the promoter of the genes of this invention; and c) identifying hypomethylation of the aforementioned region(s) in the TDRD1 gene and/or hypermethylation of the region(s) of at least one, two, three, four or five genes selected from the group: RARB, GSTP1, APC, CCND2, PTGS2, BCL2, RASSF1, LGALS3, CDH13, PITX2, and HOXD3, wherein hypomethylation and/or hypermethylation on CpG and/or non-CpG dinucleotides of said gene(s) is identified as being different when to the same region(s) of the gene(s) or associated regulatory region in a subject not having the prostate cellular proliferative disorder,

wherein detection of said hypomethylation and/or hypermethylation is indicative for the stage/type or grade of the prostate cancer.

This unexpected finding allows to diagnose for hormone-independent cancers by a simple assay that detects the hypermethylated CpG islands in the promoter region or upstream of the promoter region of the genes of this invention directly by for instance restriction endonuclease analysis to select the proper treatment for subjects with a prostate cancer, depending on the fact of the prostate cancer is hormone refractory or hormone sensitive or depending on the stage or grade of prostate cancer as can be indicated by the hypermethylation status. This is more reliable than detecting levels of mRNA or gene products of said genes. The diagnostic methods will also allow to indicate the proper treatment for hormone -refractory cancers or avoid that subjects with an hormone sensitive cancer will receive an inadequate treatment or assure that they can be treated differently. For instance patients by the diagnosis of present invention to have hypermethylation of a CpG island in the promoter region or upstream of the promoter region of at least one, two, three, four or five genes from the group of genes: RARB, GSTP1, APC, CCND2, PTGS2, BCL2, RASSF1, LGALS3, CDH13, PITX2, and HOXD3, and in some embodiments of this invention additionally have hypermethylation of a non-CpG island in the promoter region or upstream of the promoter region of the MSMB gene, can be subjected to an antimitotic drug therapy methods of treatment or the treatment can now adequately be directed to replacing the hypermethylated CpG islands (or non-CpG islands) with a non-methylated islands which for instance is possible by a treatment with a therapeutically sufficient dosage of a pharmaceutically acceptable DNA methylation inhibitor. A particular treatment selected based on the conclusion of the diagnosis method of present invention can also be a treatment to decrease the expression of the histone modifier gene, EZH2, or a treatment to decrease the activity of the EZH2 protein. Such treatments are available in the art. For instance Chroma Therapeutics developed a series of compounds that inhibit specifically EZH2. The findings of the present invention allow to diagnose prostatic cells or tissues for prostate cancer and to distinguish between a condition of benign prostate hyperplasia and prostate cancer.

The findings of present invention now specifically allow to diagnose for androgen- independent prostate cancer (AIPC) by a simple assay that detects the hypomethylated and /or hypermethylated promoter or upstream region of the promoter directly of the genes of this invention (TDRD1 hypomethylated and RARB, GSTP1, APC, CCND2, PTGS2, BCL2, RASSFl, LGALS3, CDH13, PITX2, and HOXD3 hypermethylated) and to select the proper treatment for subjects with this AIPC or avoid that subjects with an androgen sensitive cancer will receive an inadequate treatment or allow that a such subject will be treated differently than subjects with androgen-sensitve prostate cancer.

Several documents are cited throughout the text of this specification. Each of the documents herein (including any manufacturer's specifications, instructions etc.) are hereby incorporated by reference; however, there is no admission that any document cited is indeed prior art of the present invention.

EXAMPLES

Example 1: Materials and Methods Patients and sample collection

40 benign prostate hyperplasia (BPH) samples, 69 prostate cancer (PCa) samples (PCal cohort), 67 PCa samples (PCa2 cohort) and 16 PCa samples (PCa3 cohort), all paraffin embedded, were obtained from the UZ Leuven (Leuven, Belgium). Characteristics of the cohorts of patients analyzed are represented in Table 1. PCa patients from PCal cohort were selected based on following criteria:

■ Patients < 75 years of age, with histologically proven invasive adenocarcinoma of the prostate

■ Patients with, according to the Partin tables, a risk of equal or more than 10% but less than 35% of lymph node metastasis (intermediate to high risk of locoregional or metastatic disease)

■ No involvement of pelvic lymph nodes assessed by CT scan

■ No evidence for bone metastasis ^■ WHO performance status > 2

^■ No previous pelvic irradiation or radical prostatectomy

^■ No previous hormonal therapy

^■ No other malignancy except adequately treated basal cell carcinoma of the skin or other malignancy from which the patient has been cured for at least 5 years

All patients provided written informed consent according to ICH/GCP, and national/local regulations.

Additionally BPH samples from 2 patients (analyzed together with 40 paraffin embedded BPH samples) and matched PCa/adjacent normal tissue samples from 7 patients were obtained from the University of Liege (Belgium). All tumor samples were obtained from macroscopically abnormal areas within radical prostatectomy specimens. All samples were directly obtained in the OR and snap frozen in liquid nitrogen vapors. Before snap freezing, a slice was cut for formalin fixation and further paraffin embedding. Sections were cut from these slices and the percentage of tumor was estimated (tumor samples contained at least 80% of tumor cells).

In addition, the prostate cell lines LNCaP, DU 145, PC-3, PZ-HPV-7, BPH1 (American Type Culture Collection, Rockville, MD, USA) and human genomic DNA (Clontech Laboratories, Inc., Mountain View, CA, USA) were used in the experiments.

Nucleic acid extraction

Genomic DNA was extracted using the GenElute Mammalian Genomic DNA Purification Kit (Sigma-Aldrich, St. Louis, MO, USA) for cell lines and snap-forzen tissues, and the WaxFreeTM DNA kit (TrimGen, Sparks, MD, USA) for paraffin-embedded tissues following the manufacturer's protocol. The concentration of DNA was determined with the spectrophotometer NanoDrop ND-1000 (Thermo Fisher Scientific, Wilmington, DE, USA). Methylation analysis

Genomic DNA from all prostrate samples (500 ng) was bisulfite-converted using the EZ DNA methylation kit (Zymo Research Corp., Orange, CA, USA) according to the manufacturer's protocol. The final elution of bisulfite treated DNA was done in 25 ul elution buffer. Samples were stored at -80°C. The modified DNA was used as a template for quantitative multiplex nested-MSP.

Quantitative multiplex nested-MSP analysis

Quantitative multiplex nested MSP analysis was performed in two subsequent steps. In step 1, multiplex nested PCR was performed to co-amplify 12 genes, using external primer pairs independent of DNA methylation, i.e. containing no CpG sites, or no more than one CpG site close to 5' end, designed according to guidelines (MSP PCR, PCR11). All primers are listed in Table 2. PCR was performed in a volume of 25 ul containing reaction buffer (16.6 mM (NH₄)₂S0₄, 67.0 mM Tris pH 8.8, 6.7 mM MgCl₂x6H₂0, 10.0 mM β-mercapto-ethanol), 2,5 ul of dNTP Mix 2mM each (Fermentas GmbH, St. Leon-Rot, Germany), 2.5 ul of 10* 24 primer mix 2uM each primer (Sigma-Aldrich N.V. Bornem, Belgium), 0.5U IMMOLASE™ DNA polymerase (Bioline USA Inc., Boston, MA, USA), 3 ul of bisulfite-converted DNA template. Reactions were carried out in triplicate using the following conditions: 95°C for 10 min, then 30 cycles at 95°C for 30 s, 57°C for 30 s, 69°C for 30 s; and a final extension step at 69°C for 3 min. A negative control for the assay (water only) was included. The final PCR product from each triplicate was diluted 1 :500 in sterile distilled water. In step 2, separate quantification of methylated and unmethylated DNA fragments of each gene preamp lifted in step 1 was performed in two independent quantitative reactions (MSP and USP) containing a pair of internal primers, correspondingly, for methylated (M) or unmethylated (U) sequences, for each of 3 repeats separately, on a Rotor-Gene TM 6000 (Corbett Life Science Pty Ltd, Mortlake, NSW, Australia). Reactions were carried out in a volume of 15 ul in the same PCR mix with addition of 0.75 ul EvaGreen® dye (Biotium Inc, Hayward, CA, USA), 0.4 uM of M or U forward and reverse primers (listed in Table 2), 0.3U IMMOLASE™ DNA polymerase (Bioline, London, UK) and 2ul of diluted PCR product from multiplex nested PCR. Cycling conditions were as follows: 95°C for 10 min, then 30-35 cycles at 95°C for 20 s, 61°C for 15 s, 69°C for 15 s. Melting curve analysis of amplification products was performed at the end of each PCR reaction by increasing the temperature from 70°C to 95°C by 0.5°C every 10 s. Separate standard curves were generated for MSP and USP using 5 serial dilutions of plasmids (3x 10⁷-3^X 10³ copies per reaction in triplicate) containing a cloned fragment of each gene of interest in, correspondingly, a fully "methylated" (with Cs in CpG dinucleotides) or "unmethylated" (containing no Cs) versions, and the number of "methylated" (M) and "unmethylated" (U) molecules for each gene in every cancer sample was quantified. The percent methylation for each gene in the panel was calculated as %M = [M / (U + M)] 100.

Plasmid M and U clones were obtained by separate amplification of a promoter region of every gene with methylation independent primers using alternatively methylated PCa cell lines (M standard) and human genomic DNA from whole blood (U standard) under the PCR conditions listed above. Amplified fragments were cloned in DH5a™ competent cells (Invitrogen Ltd, Paisley, UK), using pGEM®-T Easy Vector System (Promega Corporation, Madison, WI, USA).

Statistical analysis

Using the CpG island methylation data, sensitivity, specificity, positive predictive value (PPV) and negative predictive value (NPV) of every DNA methylation marker were determined using online Sensitivity and Specificity Calculator at http://www.josephsunny.corn/medsoft/sensitivity_specificity.html based on a methylation treshold below which a sample was regarded as unmethylated. The latter was determined for each marker based on its degree of methylation in BPH tissues. The methylation threshold was set at \%CCND2, 2% for RARB, GSTP1, APC, PTGS2 and BCL2, 5% for TDRD1, 15% for LGALS3, PITX2 and CDH13, 20% for RASSF1 and HOXD3. To determine the relation between methylation and clinicopathologic characteristics, the second methylation threshold, or a cutoff value (CV), was introduced for all markers based on the median methylation value (MV), which was calculated for samples methylated > 1%. If for some gene the median value was lower than the first methylation threshold determined based on methylation of BPH, the latter was applied as a cutoff value. For the PCal cohort the following statistical analysis was performed: using the methylation value (MV) at each gene as a dichotomous independent variable with MV > CV for each gene designated as highly methylated (HM) and MV < CV designated as lowly methylated (LM), the relation between the high/low DNA methylation and a categorical clinicopathologic variable was assessed by means of a Fisher's Exact test. Categorical clinicopathologic variables included pT stage (I, II vs III, IV), Gleason score (4- 7a vs 7b- 10) and lymph nodes (negative vs positive). For continuous clinicopatho logic variables (PSA level, tumor volume, age of patients), the mean variable was described descriptively between the samples with and without methylation and compared between the two groups by means of a Wilcoxon-rank sum test. Graphs were generated using GraphPad Prism 3.0 software (GraphPad Software, Inc., La Jolla, CA, USA).

For the PCa2 cohort similar statistical methodology with some variations was applied. Namely, the univariable association between two continuous variables or a continuous and ordered categorical variable was analyzed by the Spearman correlation coefficient. The association between a continuous and a categorical variable was analyzed by means of the Mann- Whitney U test (for two categories) or the Kruskal-Wallis test (for more than two categories). The association between two categorical outcomes was analyzed by means of the chi-square test or the Fisher exact test (in case of low (<5) cell frequencies).

For studying the relationship between DNA methylation percentage and time-to-event outcomes (BCR and clinical failure) a Cox proportional hazard model was used. The presence of a nonlinear relationship between the methylation percentage and event risk was investigated by considering models with a quadratic effect and using restricted cubic splines. The fit of these models was compared to the fit of a model with linear trend using a likelihood ratio test. Based on the best fitting model proposals for categorizations of the DNA methylation percentage were derived. In case of a linear trend the cutoff value that leads to the best dichotomization (highest likelihood) was selected by comparing all possible dichotomisations. Further the predictive accuracy of competing models (e.g. with continuous as categorical versions of the DNA methylation percentage) by means of the concordance probability index (CPE; Gonen and Heller 2005) was calculated. A graphical presentation of the results is given by plotting the Kaplan Meier estimates in case of categorical predictors or the model-predicted incidence curves for a selection of predictor values in case of continuous predictors.

The association between the methylation in primary tumor and metastasis for every single gene in the PCa3 cohort was analyzed by means of the Pearson correlation coefficient. As a global measure of correlation for all genes, the canonical correlation coefficient was calculated.

All analyses have been performed using SAS software, version 9.2 of the SAS System for Windows. Copyright © 2002 SAS Institute Inc. SAS and all other SAS Institute Inc. product or service names are registered trademarks or trademarks of SAS Institute Inc., Cary, NC, USA. Example 2: Results

1. Screening of prostate cancer-related hyper- and hypomethylated genes

The average number of genes methylated with functional significance in PCa is estimated as several hundred (1). Kim et al (2) identified 812 out of 1171 unique gene promoters methylated only in prostate cancer cell line LNCaP in comparison with the benign cell line PrEC. We started from the hypothesis, however, that all these methylation changes are the consequence of a general failure of the DNA methylation machinery in the onset of tumorigenesis and during cancer progression, and targeting of a small number of differentially methylated sites is quite enough to develop a methylation assay with substantial diagnostic and prognostic significance. Therefore our aim was to select from a pool of the described in the literature methylation markers of PCa a limited number of genes showing altered methylation status between normal and tumor tissue, and in a perspective - between different stages of PCa. For this purpose we screened a selection of genes reported to be frequently differentially methylated in PCa and benign prostate tissues and/or associated with PCa prognosis and outcome (2, 3 and other) using a set of 6 model genotypes. These genotypes comprised three PCa cell lines (androgene-sensitive LNCaP, androgene-refractory PC-3 and DU 145, corresponding to early and late stage cancer) and two benign prostate cell lines PZ- HPV 7 and BPHl, as well as a genomic DNA sample extracted from the whole blood, which corresponds to non-cancerous DNA with regular levels of methylation (Fig. 1A). For the Melting curve analysis methylation independent primers were designed to amplify promoter fragments of the candidate genes and analyze their methylation status by the melting curve assay to detect the presence of unmethylated and partially or fully methylated copies (Fig. IB). The analyzed genes are classified into two groups: hyper- and hypomethylated markers. CDH1, SOCS3 and MIR132, which are reported to be hypermethylated in PCa, showed no sign of methylation in a group of our model genotypes (Table 3). RARRES1, which is frequently methylated in PCa based on literature, showed similar levels of methylation in both benign and cancer genotypes except in LNCaP, where it was heavily methylated. KRT7 and TACSTD2 are the new PCa cancer markers recently identified in a study of the global reactivation of epigenetically silenced genes by treatment prostate cell lines with 5Aza-dC (1). In our experiment KRT7 was methylated in one benign and one cancer genotype, TACSTD2 was methylated in one cancer genotype.

Among the hypomethylated markers SNCG reported to be hypomethylated in metastatic PCa tumors surprisingly was methylated only in cancer cell lines. The same has been observed for another hypomethylation marker MMP2. Hypomethylation marker HPSE was unmethylated in all model genotypes, while TDRD1 was fully methylated. TFF1 showed a significant reduction of the methylation levels in one benign and two PCa cell lines. The MAGEA2 promoter region was completely methylated in all genotypes except LNCaP cells, where it was completely unmethylated (Fig. 2). Conclusion

We have selected 38 genes linked to PCa from the literature (Table 3) and analysed their DNA methylation state in three PCa and two benign prostatic cell lines, and human whole blood by melting curve analysis. Based on the semi-quantitative data of the melting curve analysis we selected candidates for further analysis of their role in PCa development. The selected group contained four types of biomarkers: 1) genes frequently hypermethylated in PCa (diagnostic markers with possible prognostic significance) - APC,BCL2, ELF4, GSTP1, HAAO, HOXD3, LGALS3, PTGS2, RARB, RASSF1, RGS22, SHOX2, SOX1, TBX20, TDRD5; 2) metastatic marker: CFTR, NLGN1, PTF1A; 3) markers associated with biochemical recurrence after prostatectomy AIMF2, CHST7, APC; 4) hypomethylation markers TDRD1, MAGEA2. From this group we selected 16 genes and developed a two-step quantitative multiplex nested-MSP as detailed below. 2. Development and validation of the two-step quantitative multiplex nested-MSP

Methylation-independent primers either from the screening step or designed separately were used to amplify a part of a CpG island in the promoter regions of the selected genes. PCR amplification was performed separately for each gene using genotypes that had shown differential methylation of a separate gene by melting curve analysis. Subsequently, the PCR fragments were subcloned in pGEM-T-easy plasmid vector and multiple clones were sequenced. The sequence information was used to validate the correct amplification of the gene and to determine the methylation status of all CG dinucleotides. The plasmids, containing the PCR fragment corresponding to the fully methylated and unmethylated DNA, were selected and labeled as plasmid M and U, respectively. After obtaining M and U standards for every gene, a final set of methylation independent primers was designed for amplification of PCR fragments around 100 base pairs (bp) in length whenever possible, but not exceeding 200 bp (listed in Table 1). Also, two sets of nested primers specific for either methylated or unmethylated bisulphite modified DNA sequence of each gene were designed (Table 1). To examine the specificity of the nested methylation and unmethylation specific primers, both plasmids M and U were used as a template for amplification with both primer sets (Fig. 3). Importantly, no methylated signal (0 % methylation) was detected with plasmid U, as well as in "no template" control, using methylation specific primers (qMPS reaction). Likewise, no unmethylated signal was detected with plasmid M, as well as in "no template" control, using unmethylated specific primers (qUSP reaction), indicating that the qMSP and qUSP reactions were 100% specific.

To check the efficiency of both sets of primers, separate standard curves were generated for MSP and USP using 5 serial dilutions of plasmid standards M and U (3>< 10⁷-3x l0³ copies per reaction in triplicate) for each gene. Final sets of qMSP and qUSP amplifying the template with high efficiency and linearity were selected (Fig. 4). The whole scheme of the validation process is presented on figure 5. The scheme of the two-step quantitative multiplex nested- MSP is presented on figure 6. Briefly, a mixture of validated gene-specific multiplex primers was used to co-amplify 12 gene promoters independent of their methylation status. The amplification product was diluted 1 :500 with sterile distilled water and used as a template for the quantitative PCR. In the second PCR reaction, absolute quantification of methylated and unmethylated DNA fragments for each gene was performed separately by using validated M and U specific primers and separate standard curves generated using serially diluted M and U plasmid standards.

Conclusion. The advantage of the developed two-step quantitative multiplex nested-MSP assay is that it utilizes the same bisulphite-converted DNA template (which is often very limited in a volume and quantity) to preamplify the selected number of gene promoters of interest in one PCR tube. This makes the procedure independent of sampling and pipeting diversions and allows at the same time to obtain the sufficient amount of the DNA targets for MSP primers to reduce false priming errors. The second quantitative step enables (the researcher) to determine the lowest methylation levels and discriminate between functionally significant and insignificant or background methylation.

3. Determination of DNA methylation state of 16 genes in prostate cell lines and paired PCa/adjacent normal tissue samples

The developed two-step quantitative multiplex nested-MSP assay was used to determine the degree of methylation of the selected genes in prostate cell lines as well as in PCa samples and BPH genotypes. As it has been observed in a previous study (4), cancer lines usually exhibit higher levels of CpG island hypermethylation than primary cancers, which may be a result of repeated passages and adaptation to culture environment, as well as of contamination of tumor samples by adjacent normal cells. In our study PCa cell cultures also show more polar methylation values as compared to the PCa samples (Fig. 7 and 8). LNCaP cell line corresponding to hormone-dependent (early stage) PCa surprisingly showed the highest methylation value for 11 out of 14 hypermethylated genes analyzed. The hypomethylation marker TDRD1 was also 100% methylated in LNCap, but the DNA methylation state of TDRD1 is indicated as a reverse value (i.e. 100 - % of methylation corresponding to unmethylation frequency). However, two markers of PCa progression and biochemical recurrence CDH13 and HOXD3 were not significantly methylated in this line. PC-3 showed the highest value for 3 genes reported to be associated with biochemical recurrence: APC, PITX2 and HOXD3, DU 145 - for HOXD3 and CDH13. A combination of these markers has a greater prognostic value in comparison with that of a single marker.

Low levels of GSTP1 methylation in advanced PCa cell lines PC-3 and DU 145 may be a result of hypomethylation processes active at the later PCa stages, as well as the absence of LGALS3 methylation, which frequently methylated on the onset of tumorigenesis (and is 100% methylated in LNCaP), but is reported to become heavily methylated in earlier PCa stages, but is only lightly methylated in more advanced stages III and IV (5). In general these two cell lines had lower number of genes methylated above 90% (9 in PC-3 and 7 in Du 145) as compared with LNCaP cells. Among non-malignant cells both BPH 1 and PZ-HPV7 showed methylation higher than 10 % for 5 different genes each (Fig. 7). In benign genotypes methylation higher than 20% was detected for RASSF1, PITX2, HOXD3, TDRD5, TBX20 and SOX1 implying that a higher methylation cutoff value must be introduced for these markers to reveal their diagnostic and/or prognostic significance. Apart from the hypomethylation markers TDRD1 and MAGEA2, only two genes were methylated higher than 2% in HG DNA: TDRD5 (14.16%) and SOX1 (3.71%).

Methylation of all genes was detected to a much greater extent in tumor samples in comparison with histologically cancer-free adjacent tissues (Fig. 8). However, methylation of at least one of three PCa-specific genes GSTP1, RARB and APC was detected above 2 % methylation threshold in all adjacent normal tissues (except a specific case of sample 2). This is a result of so-called "field methylation" phenomenon, detected in many studies of paired tumor/adjacent normal tissue studies, which proves that the utilization of methylation markers may be useful for the detection of PCa in biopsies that are histologically tumor-free (false- negative biopsies), reducing the need for repeat biopsies.

The PCa samples 1 and 6 have a higher Gleason score 3+4 compared to other samples (GS 3+3). In PCa sample 1 and 6, 9 and 6 genes respectively, show methylation close to or above the 50% level. PCa samples 5 and 7 show lower levels of methylation, which corresponds to the notion that the degree of methylation increases with tumor progression. However, the PCa samples 3 and 4 have, 7 and 8 genes hypermethylated, respectively, close to or above 50%; and in addition, only these samples show significant hypomethylation of the TDRD1 gene, distinguishing PCa 3 and 4 from the other PCa samples. Taken together, this suggests that samples PCa3 and 4 are another type of PCa than the other PCa samples. Finally, HOXD3 shows methylation above median in a sample PCa 4 and PCa 6, while PITX2 has a higher methylation level in sample PCa 1. Sample PCa 2 shows very low levels of hypermethylation of PCa-specific genes (GSTP1, RARB, APC, CCND2 and PTGS2) and no methylation of those genes in the paired benign tissue, suggesting that hypermethylation is not really involved in the pathogenesis of that tumor. However, the benign biopsy shows elevated levels of methylation of RASSF1, LGALS3 and CDH13 in comparison with the cancer tissue. Although these markers are often moderately methylated in benign or normal prostate tissue, their methylation most often significantly increases in tumor, in contrast to the methylation pattern in a PCa sample 2. Taken together, this suggests the involvement of DNA demethylation processes, which keeps PCa-specific genes from getting hypermethylated and affects DNA targets methylated at earlier stages. The hypomethylation marker TDRD1, however, does not seem to be affected by demethylation. This specific cancer case (sample PCa2) signalizes that either not all PCa cases are followed by substantial DNA hypermethylation changes or these changes may be initiated, but suppressed and kept at a quite low level by a certain unknown mechanism. An alternative explanation is that in some tumors the waves of hypomethylation during cancer progression may affect the DNA targets hypermethylated at earlier stages, thus reducing the levels of methylation of some established hypermethylation markers. Based on the two-step quantitative nested-MSP, we identify sample PCa2 as a methylation-independent cancer case.

Conclusions

1. The developed two-step quantitative multiplex nested-MSP assay effectively distinguished PCa cell lines from non-malignant cells, as well as PCa tumors from surrounding malignant tissues based on the quantification of the methylation values of 16 markers. 2. Cancer cell lines exhibited higher methylation values compared with the PCa samples. LNCaP cells showed the highest number of completely methylated genes (90-100%): 11 out of 14 hypermethylated markers.

3. The two-step quantitative multiplex nested-MSP assay effectively detected low methylation values of PCa-specific methylation markers in non-malignant biopsies from PCa patients due to a phenomenon called "methylation field effect". This implies that the assay has a power to detect PCa in histologically negative biopsies (false-negative biopsies).

4. A PCa2 sample showed very low methylation levels of cancer-specific markers, while methylation values of some other markers were even lower in the tumor than in the surrounding non-malignant tissue. We classify this prostate cancer as methylation- independent PCa type. Differential methylation in such tumors may not occur or hypermethylation is suppressed and/or kept at a very low level due to demethylation processes or due to some other unknown mechanisms.

4. Determination of DNA methylation state of 12 genes in benign prostate hyperplasia and PCa samples

To determine sensitivity and specificity of the methylation markers, we applied the developed two-step quantitative multiplex nested-MSP assay for quantification the levels of methylation of 12 genes in four groups of patients: BPH (benign prostate hyperplasia, N=42), PCal (N=69), PCa2 (N=67), PCa3 (N=16). Based on the methylation levels in the BPH cohort (Table 4), we divide the analyzed genes into two groups: diagnostic and prognostic markers. Diagnostic markers, also indicated as PCa-specific markers, have no or very low levels of methylation in BPH samples not exceeding 1-2% of methylated gene copies. Thus, at setting the methylation cutoff value at 2% such markers attain 100% specificity, showing at the same time quite high sensitivity (Table 5). The group of PCa-specific markers includes RARB, GSTPl, CCND2, PTGS2, BCL2 (all 5 show 100% specificity at the 2% methylation cutoff value) and APC (98%> specificity). The second group of genes was moderately methylated in BPH samples, so the methylation cutoff value should be raised to 5-20% of methylated gene copies to increase specificity of these markers. This decreased their sensitivity, while specificity still never reached 100% (Table 5). However, we will utilize these markers for prognostic rather than diagnostic purposes. The group of prognostic markers includes LGALS3, TDRD1, RASSF1, PITX2, HOXD3, CDH13.

DNA of patients from the PCal cohort was extracted from the whole paraffin slide containing both tumors and non-tumor cells, while in the PCa2 cohort it was extracted exclusively from tumor sites. This explains why methylation values were lower in the PCal cohort compared with that in the PCa2 cohort for most of the genes (Table 4), as the tumor DNA in the PCal cohort was mixed with the DNA from surrounding non-malignant cells. This also explains why sensitivity of most of the markers was lower in the PCal cohort (Table 5). Methylation of markers could be detected, but due to presence of the unmethylated DNA from non- malignant cells the detected methylation levels could be lower than the methylation cutoff set for each separate marker (1-2% methylation for PCa-specific markers, 5-20% for prognostic markers.

In general, and considering the all said above, the two-step quantitative multiplex nested-MSP assay detected PCa cancer with very high efficiency. In the PCal cohort only 5 of 69 samples showed no significant methylation of PCa-specific genes. In this group a combination of GSTPI + RARB + APC + BCL2 had highest sensitivity of 92.75% at 100% specificity. In the PCa2 cohort the combination of RARB + GSTPI had sensitivity 100.00% at 100% specificity, while sensitivity of GSTPI alone was 96.97%. We attribute the cases with the insignificant methylation values detected in the PCal cohort to methylation-independent PCa types, having, however, quite low incidence. Still there is a number of cases in every cohort with methylation above the cutoff value, but not exceeding 10%> for most of the genes (5.79% and 10.77%) of cancer cases in the PCal and PCa2 cohorts correspondingly). We would prefer to call these cases as tumors with suppressed methylation; this phenomenon may have an impact on the prognosis of the tumor development based on methylation marker information and needs further more detailed investigation. Conclusions

1. The two-step quantitative multiplex nested-MSP assay showed very high sensitivity for PCa at the 100% with specificity level: 92.75% in the PCal cohort (mixed DNA from malignant and non-malignant cells) and 100.00% in the PCa2 cohort (DNA predominantly from the tumor sites). 2. 7.25% of tumors in the PCal cohort may be attributed to methylation-independent cancer cases, as none of the markers (or no more than one marker) showed methylation above the cutoff value in such tumors. Methylation-independent tumors may be identified by the two-step quantitative multiplex nested-MSP assay and discriminated from methylation- associated tumors in case if PCa was detected by other diagnostic means. 3. Tumors with low (below 10%) methylation of all markers were detected in all two PCa cohorts analyzed; the incidence of such cases was 5.79% in PCal and 10.77% in PCa2. The two-step quantitative multiplex nested-MSP assay is able to detect such methylation- suppressed PCa tumors; however, no prognostic conclusion may probably be made for such cancer cases based on DNA methylation, as the methylation seems to be suppressed by some unknown mechanism.

5. Correlation of CpG island methylation alterations with clinicopathologic parameters. Correlations of CpG island methylation alterations with clinicopathologic parameters were analyzed separately in the PCal and PCa2 cohorts because the DNA samples from these cohorts originated from different sources (mixture of tumor and non-tumor cells in PCal and just tumor cells in PCa2).

We evaluated correlations between CpG island alterations of ten genes (single loci, RARB, GSTP1, CCND2, PTGS2, APC, LGALS3, TDRD1, RASSF1, CDH13, BCL2) and clinicopathologic parameters: preoperative serum PSA level, clinical tumor pT stage, Gleason score, lymph node involvement, tumor volume, age in the PCal cohort (Tables 6-7). In this cohort the univariable association of DNA methylation with clinical tumor pT stage, Gleason score and lymph node metastases was measured as the association between a continuos (% of DNA methylation) and a categorical variable (pT stage 1-2 vs 3-4; Gleason score 4-7a vs 7b- 10; lymph nodes positive vs negative) by a Wilcoxon-rank sum tests. The univariable association of DNA methylation with the preoperative PSA level, age of the patients and tumor volume was measured as the association between a categorical (low or high DNA methylation) and a continuous variable (PSA, age, tumor volume) by means of Fisher's exact test. RARB, RASSF1 and CDH13 methylation showed association with tumor stage; PTGS2 and RASSF1 methylation, respectively, with Gleason score and lymph node involvement. CpG island hypermethylation at GSTP1, BCL2 and RASSF1 strongly correlated with tumor volume. Apart from a strong association with tumor volume the only hypomethylation marker TDRD1 also showed association with pT stage. There was no correlation between gene methylation alterations and age and preoperative PSA in the PCal cohort. Level of DNA methylation of APC, CCND2 and LGALS3 showed no correlation with clinicopathologic parameters.

We also evaluated correlations between CpG island alterations of eleven genes (single loci, RARB, GSTP1, CCND2, PTGS2, APC, LGALS3, TDRD1, RASSFl, PITX2, HOXD3, CDH13,) and clinicopathologic parameters: preoperative serum PSA level, clinical tumor pT stage, extracapsular extension, seminal vesicle invasion, surgical margin status, Gleason score, age and tumor volume in the PCa2 cohort (Fig. 9, Table 8).

No correlation between DNA methylation and preoperative PSA levels was found, when those variables were regarded as continuous variables (analyzed by the Spearman correlation coefficient). However, the association between PSA and DNA methylation of GSTPl and RASSFl was found in case of dichotomization of DNA methylation (low vs high methylation, Mann- Whitney U test, Table 9).

DNA methylation was not associated with the age of patients (Spearman correlation, Table 9). Methylation of PITX2 and HOXD3 was highly associated with pathological tumor pT stage (pT as unordered categorical variable: pTl-2, 3a, 3b, 4; DNA methylation as a continuous variable, Kruskal- Wallis test, Table 8) and ecstracapsular extention (Mann-Whitney U test, Table 8). Besides, HOXD3 methylation was associated with seminal vesicle invasion (Mann- Whitney U test, Fisher exact test, Table 8) and lymph node invasion (Mann- Whitney U test, Table 9), although the number of patients with lymph node metastases in the PCa2 cohort was rather low (7 of 67). Similar to the PCal cohort, RARB and TDRDI showed association with tumor volume (measured as a percentage of a total gland volume) in the PCa2 cohort. The same association was detected also for HOXD3 (Mann-Whitney U test, Fig. 9). In case of measuring the tumor volume in mL, an association of this variable with methylation of CCND2, PTGS2, APC and RASSF1 was also detected (Fig. 9). CCND2 showed association with Gleason score (Gleason score 2-6, 7, 8-10, Kruskal- Wallis test, Table 8). All statistically significant association between clinicopathologic parameters and DNA methylation at least in one of the two PCa cohorts analyzed are indicated in Table 9. Conclusion

Some significant associations between CpG island methylation levels and clinicopathologic parameters are detected. DNA methylation state of TDRDl and RARB showed strong correlation with tumor volume in both PCa cohorts analyzed.

Example 3: Results on prognostic significance of DNA methylation markers.

To assess the prognostic value of DNA methylation markers, we analysed methylation frequencies in a PCa2 cohort of 69 patients with < 16 years of follow-up data. The primary clinical end point of this study was biochemical recurrence (BCR, also biochemical relapse or biochemical progression- free survival). Clinical failure was the second end point of this study. BCR was detected in 37 (55.22%) patients and for 11 (16.42%) patients clinical failure was observed. PCa samples were divided into high methylation (HM) and low methylation (LM) groups based on methylation cutoff value selected for each gene individually (Table 10).

Univariate Kaplan-Meier curves demonstrated that low-level methylation of TDRDl, and high high-level methylation RASSFI, PITX2 and HOXD3 predicted significantly earlier biochemical recurrence with the best methylation cutoff value at 69. 23% percentile, 71.88% percentile, 72.00%) percentile and 35.82%> percentile correspondingly (Fig. 10-11). High PITX2 methylation was also associated with the significant risk of clinical failure (CF, methylation cutoff value at 93.94% percentile, Fig. 12). Multivariate analysis of the association between combinations of the TDRD1, RASSF1, PITX2, HOXD3 genes and clinical outcomes demonstrated that a combination of these four markers predicts the risk of BCR with the higher predictive accuracy (concordance probability estimate, CPE) compared with the monomarker models (Table 11).

None of the methylation markers highly specific for PCa (RARB, GSTP1, CCND2, PTGS2, APC) showed significant prognostic value. This observation allows to separate methylation markers analyzed in this study into two groups: diagnostic markers {RARB, GSTP1, CCND2, PTGS2, APC) and prognostic markers (TDRD1, RASSF1, PITX2, HOXD3). Conclusion

Low methylation levels of TDRD1 and high methylation levels RASSFI and HOXD3 are significantly associated with the risk of BCR, and high methylation of PITX2 is associated with the risk of both BCR and CF. A multivariate models comprising TDRD1, RASSFI, PITX2 and HOXD3 predicts the risk of BCR with higher accuracy.

Example 4: Correlations between DNA methylation in primary tumors and metastases.

Correlation between DNA methylation frequencies of 11 genes in primary tumor sites and lymph node metastatic sites from the same patients (one tumor DNA sample was compared to one lymph node site for each patient) was analyzed in the PCa3 cohort (N=16). Pearson correlation coefficient was higher than 0.8 for CCND2, PTGS2, TDRD1, higher than 0.6 for RARB, APC, PITX2, HOXD3, higher than 0.4 for GSTPI, LGALS3, RASSFI and CDH13 (Table 12). Maximal canonical correlation for the combination of 7 genes with higher Spearman correlation coefficients determined by a univariate analysis was 0.9989.

Conclusion

High correlation of the DNA methylation levels in primary tumor and lymph node metastatic sites was detected for the most of genes analyzed. This proves the validity of the developed two-step quantitative multiplex nested-MSP assay.

Literature

1. Ibragimova I, Ibanez de Caceres I, Hoffman AM, Potapova A, Dulaimi E, Al-Saleem T, Hudes GR, Ochs MF, Cairns P. Global reactivation of epigenetically silenced genes in prostate cancer. Cancer Prev Res (Phila). 2010;3(9): 1084-92. 2. Kim JH, Dhanasekaran SM, Prensner JR, Cao X, Robinson D, Kalyana-Sundaram S, Huang C, Shankar S, Jing X, Iyer M, Hu M, Sam L, Grasso C, Maher CA, Palanisamy N, Mehra R, Kominsky HD, Siddiqui J, Yu J, Qin ZS, Chinnaiyan AM. Deep sequencing reveals distinct patterns of DNA methylation in prostate cancer. Genome Res. 2011;21(7): 1028-41.

3. Mahapatra S, Klee EW, Young CY, Sun Z, Jimenez RE, Klee GG, Tindall DJ, Donkena KV. Global methylation profiling for risk prediction of prostate cancer. Clin Cancer

Res. 2012 May 15;18(10):2882-95.

4. Chung W, Kwabi-Addo B, Ittmann M, Jelinek J, Shen L, Yu Y, Issa JP. Identification of novel tumor markers in prostate, colon and breast cancer by unbiased methylation profiling. PLoS One. 2008 Apr 30;3(4):e2079. 5. Ahmed H, Cappello F, Rodolico V, Vasta GR. Evidence of heavy methylation in the galectin 3 promoter in early stages of prostate adenocarcinoma: development and validation of a methylated marker for early diagnosis of prostate cancer. Transl Oncol. 2009;2(3): 146- 56.

Claims

2. The method according to claim 1, further comprising

3. The method according to anyone of claims 1 or 2, further comprising

a. analyzing the level of DNA methylation of the regulatory region surrounding the transcription start site (TSS) of at least one gene selected from RARB, GSTP1, APC, CCND2, PTGS2, BCL2, LGALS3, and CDH13 in said test sample; and b. comparing the methylation level of CpG dinucleotides in the regulatory region surrounding the transcription start sites of the at least one gene of step (a) in the test sample with said level in a reference sample;

wherein the methylation level of CpG dinucleotides in said regulatory region in said at least one gene selected from RARB, GSTP1, APC, CCND2, PTGS2, BCL2, LGALS3, and CDH13 is further indicative of the type and/or stage of said prostate cell proliferative disorder.

4. The method according to anyone of claims 1 to 3, wherein hypomethylation of CpG dinucleotides in the regulatory region surrounding the transcription start sites of TDPvDl is correlated to a prostate cell proliferative disorder of a more aggressive type and/or of a further advanced stage.

5. The method according to claim 2, wherein hypermethylation of CpG dinucleotides in the regulatory region surrounding the transcription start sites of the at least one gene selected from PITX2, RASSF1, and HOXD3 is further correlated to a prostate cell proliferative disorder of a more aggressive type and/or of a further advanced stage.

6. The method according to claim 3, wherein hypermethylation of CpG dinucleotides in the regulatory region surrounding the transcription start sites of the at least one gene selected from RARB, GSTP1, APC, CCND2, PTGS2, BCL2, LGALS3, and CDH13 is further correlated to a prostate cell proliferative disorder of a more aggressive type and/or of a further advanced stage.

7. The method according to anyone of claims 4 to 6 wherein said prostate cell proliferative disorder of a more aggressive type is a biochemical recurrent type of disorder.

8. The method according to anyone of claims 1 to 6 wherein said prostate cell proliferative disorder of a more advanced stage is a high clinical stage disorder of pT stage III or IV.

9. The method according to anyone of claims 2 or 3, wherein iso-methylation of CpG dinucleotides in the regulatory region surrounding the transcription start sites of TDRD1 in combination with iso-methylation or hypomethylation of at least one gene selected from PITX2, RASSF1, HOXD3, RARB, GSTP1, APC, CCND2, PTGS2, BCL2, LGALS3, and CDH13; in particular at least one gene selected from GSTP1, RARB, PITX2 and HOXD3, or at least one gene selected from RASSF1, LGALS3 and CDH13; is indicative of a methylation-independent type of prostate cell proliferative disorder.

10. The method of claim 1, further comprising

wherein decreased methylation of CpG dinucleotides in said regulatory region in the TDRDl gene is indicative of an hormone refractory prostate cancer type, androgen-independent prostate cancer (AIPC) type or androgen-independent metastatic prostate cancer type.

11. The method of claim 10, further comprising:

wherein increased methylation of CpG dinucleotides in said regulatory region of said gene(s) is further indicative of an hormone refractory prostate cancer type, androgen-independent prostate cancer (AIPC) type or androgen-independent metastatic prostate cancer type.

12. The method of any of claims 1 to 11, for use in deciding on the proper treatment or proper medicament dependent on the type and/or stage of said prostate cell proliferative disorder.

13. The method of claim 12, wherein detection of hypomethylation of TDRDl either or not in combination with hypermethylation of at least one gene selected from the list comprising PITX2, RASSF1 and HOXD3, RARB, GSTP1, APC, CCND2, PTGS2, BCL2, LGALS3, and CDH13 is indicative for the decision about the initiation or continuation of a treatment selected from a prostatectomy, a DNA methylation inhibitor or a compound in an effective amount to reduce male hormones.

14. The method of claim 12, wherein detection of iso-methylation of TDRD1 either or not in combination with iso-methylation or hypomethylation of at least one gene selected from the list comprising PITX2, RASSF1 and HOXD3, RARB, GSTP1, APC, CCND2, PTGS2, BCL2, LGALS3, and CDH13 is indicative for the decision about the initiation or continuation of a treatment selected from a prostatectomy, a histone deacetylase inhibitors, gonadotropin-releasing hormone agonists, neutraceuticals, or radiotherapy.

15. The method of any of claims 1 to 14, wherein methylation is determined using PCR analysis, bisulfite genomic sequencing PCR analysis, Methylation-Specific PCR analysis or an equivalent amplification technique.

16. The method of any of claims 1 to 15, wherein methylation is determined in an assay comprising primers for assessing the presence of methylation in a regulatory region surrounding the TSS of said gene(s).

17. The method of any of claims 1 to 16, wherein at least one primer of the group consisting of methylated specific primers (SEQ. ID NOs 3, 4, 9, 10, 15, 16, 19, 20, 25, 26, 31, 32, 37, 38, 43, 44, 49, 50, 55, 56, 61, 62, 67, 68, 73 and 74) and at least one primer of the group consisting of unmethylated specific primers (SEQ ID NO's 5, 6, 11, 12, 21, 22, 27, 28, 33, 34, 39, 40, 45, 46, 51, 52, 57, 58, 63, 64, 69, 70, 75 and 76) is used.

18. The method of any of claims 1 to 17, wherein the reference sample is a sample from a healthy individual or from an individual having a typical benign hyperplasia prostate.

19. The method of any of claims 1 to 18, wherein said regulatory region surrounding the TTS comprises one or more CpG islands and extends about 1.5 kb upstream to about 1.5 kb downstream from said transcription start site of said gene(s).

20. The method of any of claims 1 to 19, wherein the test and/or reference sample is selected from the list comprising prostatic tissue, prostatic fluid, seminal fluid, ejaculate, blood, urine, prostate secretions, histological slides, and paraffin-embedded tissue.

21. The method of any of claims 1 to 20, further comprising analysing the level of DNA methylation of at least one control plasmid selected from the list comprising SEQ ID N° 91 and 92 either or not in combination with analysing the level of DNA methylation of at least one control plasmid selected from the list comprising SEQ ID N° 77-90 and 93-100.

23. The kit according to claim 22, further comprising at least one primer selected from the list comprising SEQ ID N° 3, 4, 9, 10, 15, 16, 19, 20, 25, 26, 31, 32, 37, 38, 49, 50, 55, 56, 61, 62, 67, 68 73 and 74 and at least one primer from the list comprising 5, 6, 11, 12, 21, 22, 27, 28, 33, 34, 39, 40, 51, 52, 57, 58, 63, 64, 69, 70, 75 and 76.

24. The kit according to anyone of claims 22 or 23 further comprising at least one control plasmid selected from the list comprising SEQ ID N° 91 and 92 either or not in combination with at least one control plasmid selected from the list comprising SEQ ID N° 77-90 and 93-100.