FIELD OF THE INVENTION

The present invention relates to prostate cancer-associated genetic markers. More specifically, the invention relates to a method of predicting, diagnosing and/or prognosing prostate cancer, especially a clinically relevant, aggressive form of prostate cancer. The invention also relates to different uses of said method, and a test kit for use in said method.

BACKGROUND OF THE INVENTION

Prostate cancer is the second most common cancer in men worldwide.

An estimated 1.1 million men were diagnosed with prostate cancer in 2012, accounting for 15% of the cancers diagnosed in men, with almost 70% of the cases (759,000) occurring in more developed regions. With an estimated 307,000 deaths in 2012, prostate cancer is the fifth leading cause of male cancer deaths (6.6% of the total men deaths) worldwide.

Early detection and treatment can significantly improve prostate cancer survival. The most frequently used test to screen for prostate cancer, namely the prostate specific antigen (PSA) blood test, is effective in detecting early stage prostate cancer. Attempts to further increase the sensitivity and accuracy of PSA screening have led to the development of modified assays detecting alternative molecular forms of the molecule, including intact (iPSA), free PSA (fPSA) and total (tPSA). Some assays further incorporate the measurement of human kalligrein 2 (hK2). Despite the improvements in the PSA screening assays, they are not able to accurately distinguish between aggressive, potentially fatal cancers from clinical- ly insignificant cancers that would have caused no symptoms or harm in the patient. On the other hand, the PSA screening assays may also miss some harmful cancers. Universally however, PSA screening has led to over-diagnosis of prostate cancer as majority of prostate cancers progress slowly and only possess a low risk of prostate cancer-related morbidity and mortality.

Because of the lack of effective prognostic markers, most men with

PSA screen-detected prostate cancer undergo, often unnecessary, radical prostatectomy or radiotherapy that have side effects such as sexual impotence, urinary incontinence and bowel problems. Thus, prognostic factors for the severity of the disease would be highly beneficial to avoid overtreatment of the less aggressive and slowly growing forms of prostate cancer, and to specifically target the more radical forms of therapy to the aggressive forms of prostate cancer.

Genetic factors have been estimated to account for up to 58% of the prostate cancer risk, and to date over 100 susceptibility loci, each being a common variant with low penetrance, have been associated with prostate cancer. In subjects of Nordic descent, the highest genetic impact on the prostate cancer risk has been attributed to a rare recurrent missense mutation, namely rsl38213197/G84E, in HOXB13, a homeobox transcription factor gene that is important in prostate development. Accordingly, Published International Application WO 2013/070933 discloses a method of identifying subjects having said mis- sense mutation in HOXB13 as subjects having an increased propensity to develop prostate cancer. However, to again avoid overestimation of the risk of prostate cancer and at the same time raising unnecessary concern among the studied individuals, better markers are needed for more accurate prediction of the appearance of the prostate cancer and especially the more aggressive forms of prostate cancer. Discovery of such markers, genetic or other, would be highly beneficial to determine the risk of prostate cancer and especially the risk of aggressive prostate cancer in the overall population or for example in individuals who belong to families with a high incidence of prostate cancer.

Cancerous inhibitor of protein phosphatase 2A (CIP2A), also known as KIAA1524 and p90, is a recently identified human oncoprotein, whose expression is known to be increased at high frequency in most of the human cancer types, including prostate cancer, where it has been associated with poorly differentiated and high-risk tumours (Vaarala et ah, J Exp Clin Cancer Res. 2010; 29: 136). However, mechanisms inducing CIP2A expression in cancer still remain largely un- known. So far, CIP2A has been studied mainly on the expression and protein level, while only one study has concerned its potential role as a cancer susceptibility gene. In that study (Li et ah, Oncol Lett. 2012; 4:358-364), however, neither of the analysed mutations (rs2278911 and rs4855656) were found to be associated with the risk of hepatocellular carcinoma (HCC). The main finding of the study was that C carriers of rs2278911 exerted synergetic effect with hepatitis B and C virus infection on risk of HCC in Han Chinese population.

In accordance with the above, there is an identified need for predictive, diagnostic and prognostic markers that can preferentially differentiate between prostate cancers with lower clinical significance, including those that are localized and/or slowly growing, of those prostate cancers that are likely to progress to an advanced or lethal disease if left untreated and that cause most of the morbidity and mortality related to the disease. Such markers would be needed for example for directing the most intensive and generally most harmful forms of therapy to those prostate cancer patients who have the highest prostate cancer related morbidity and mortality and hence the highest need of therapy.

Furthermore, as developing new medical drugs and therapies for subjects suffering from prostate cancer is also challenged by the highly variable cancer-related morbidity and mortality, markers allowing more accurate selection of subjects with the more aggressive forms of prostate cancer for therapeutic trials are crucially needed to facilitate the development of new or alternative treatment strategies especially for the aggressive forms of the disease.

BRIEF DESCRIPTION OF THE INVENTION

In one aspect, the present invention provides a method of analyzing a subject for prostate cancer-associated markers, wherein the method comprises testing a sample obtained from said subject for HOXB13 rsl38213197 and CIP2A rs2278911, and wherein the presence of at least one HOXB13 rsl38213197 T allele and at least one CIP2A rs2278911 T allele indicates that said subject has an increased risk of having or developing prostate cancer, which in some further embodiments is aggressive prostate cancer, such as castration-resistant prostate cancer. In some still further embodiments, said increased risk of having or devel- oping prostate cancer indicates earlier than an average onset of clinically relevant prostate cancer, and in some even further embodiments said subject has an earlier diagnosis of benign prostate hyperplasia and/or is a Caucasian male.

The present method may be used for various purposes including but not limited to those selected from the group consisting of determining a subject's risk of prostate cancer, determining a subject's susceptibility to prostate cancer, determining a subject's risk of having or developing prostate cancer, diagnosing prostate cancer in a subject, prognosing prostate cancer in a subject, stratifying a subject into a high-risk or low-risk group for prostate cancer, stratifying a subject for monitoring of onset of prostate cancer, stratifying a subject for active surveil- lance for the progression of prostate cancer, stratifying a subject for radiotherapy or prostatectomy, stratifying a subject for clinical studies, and population screening for prostate cancer.

In another aspect, the present invention provides a kit for use in the present method and in any embodiments thereof. Said kit comprises at least one first binding body for testing said sample for HOXB13 rsl38213197 and at least one second binding body for testing said sample for CIP2A rs2278911.

In some embodiments of the present kit, said first binding body is for specifically determining the presence or absence of HOXB13 rsl38213197 (T) and/or said second binding body is for specifically determining the presence or absence of CIP2A rs2278911 (T). Alternatively or additionally, the kit may comprise further first binding bodies and/or further second binding bodies for specifically determining the presence or absence of HOXB13 rsl38213197 (C) and/or CIP2A rs2278911 (C), respectively.

In some further embodiments of the present kit, said first and second binding bodies are selected, independently from each other, from the group consisting of nucleic acid primers, nucleic acid probes, aptamers, antibodies or antibody fragments, peptides, receptors, and enzymes such as restriction enzymes.

Other objectives, aspects, embodiments, details and advantages of the present invention will become apparent from the following figures, detailed description, examples, and dependent claims.

DETAILED DESCRIPTION OF THE INVENTION

An object of the present invention is to provide a method for prediction, diagnostics and/or predictive prognostics of prostate cancer, especially ag- gressive prostate cancer.

This object may be achieved, at least partly, by a method of determining the presence of two germline mutations, namely HOXB13 rsl38213197 (T) and CIP2A rs2278911 (T), which possess a surprising synergistic effect in the increased occurrence of prostate cancer, especially aggressive prostate cancer, in subjects bearing the two mutations simultaneously. The synergistic effect utilized herein manifests itself in that the predictive, diagnostic and/or prognostic potential of the markers when occurring simultaneously is better than the sum of the parts. While the HOXB13 rsl38213197 (T) has predictive, diagnostic and/or prognostic potential (although to a lesser extent) when used alone, the CIP2A rs2278911 (T) has no such potential when used as a sole marker. Such synergistic markers are considerably more difficult to find compared to markers with direct and predictable effect on occurrence or severity of a disease and are, accordingly, typically discovered accidentally.

As used herein, the term "prostate cancer" refers to any cancer of the prostate gland, small walnut-shaped organ that produces the seminal fluid that nourishes and transports sperm, in which cells of the prostate mutate and begin to multiply out of control. The term includes not only adenocarcinomas, which develop from the gland cells and are the most common types of prostate cancer, but also rare prostate cancers including, but not limited to, sarcomas, small cell carcinomas and other neuroendocrine tumours, and transitional cell carcinomas. The term "prostate cancer" also includes all stages, categories, subcategories, and stage groupings of prostate cancer regardless of the staging system used. As way of example, the most common staging system for prostate cancer is the TNM system. The TNM system is based on the size and/or extent (reach) of the primary tumor (accounts for the letter "T" in TNM), the amount of spread to nearby lymph nodes (accounts for the letter "N"), and the presence of metastasis (accounts for the letter "M") or secondary tumors formed by the spread of cancer cells to other parts of the body.

Prostate cancer typically grows slowly and initially remains confined to the prostate gland, where it may not cause serious clinical harm. While the slowly growing types of prostate cancer may need minimal or no treatment, some other types of prostate cancer are more aggressive and can spread to other body parts, i.e. metastasize.

As used herein, the terms "aggressive prostate cancer" and "aggressive form of prostate cancer" are interchangeable and mean prostate cancers that are related with a higher cancer-related morbidity and/or mortality compared with the slowly growing, localized forms of the cancer. Characteristically the aggressive forms of prostate cancer grow faster and/or have or will develop the ability to metastasize. Examples of aggressive prostate cancers also include, but are not limited to, castration-resistant prostate cancers, poorly differentiated prostate cancers, high grade prostate cancers, and prostate cancers having a Gleason score of 8 or greater.

Since prostate cancer growth is often driven by male sex hormones called androgens, which include testosterone, a common treatment option for prostate cancer is to lower the levels of androgens in a man's body by chemical or surgical means. This type of treatment may be called hormone therapy, androgen- deprivation therapy, or castration treatment. Eventually, many metastatic prostate cancers become resistant to treatment and resume growth despite castration treatment. Prostate cancer, which is no longer responsive to castration treatment, is herein referred to by the term "castration-resistant prostate cancer" (CRPC).

As used herein, the term "screen-detected prostate cancer" refers to prostate cancer, which has been detected through increased blood level of prostate-specific antigen (PSA). Many of the screen-detected prostate cancers are clinically insignificant and would have remained undetected during life in the absence of screening. Thus, the PSA screening may lead to over-diagnosis of pros- tate cancer.

As used herein, the term "clinically-detected prostate cancer" refers to a prostate cancer which has been detected through clinical symptoms and hence indicating cancer-related morbidity. The terms "clinically significant prostate cancer" and "clinically relevant prostate cancer" are interchangeable and, in addi- tion to clinically-detected prostate cancers, also comprise aggressive, high-grade, large-volume, and/or non-organ defined prostate cancers which have progressed or are likely to progress to an advanced or lethal disease if left untreated, as well as prostate cancers related to significant morbidity or mortality of the disease.

As used herein, the term "benign prostatic hyperplasia" (BPH), also known as benign prostatic hypertrophy, refers to a common noncancerous enlargement of the prostate gland in aging men. Generally, BPH is considered neither as a premalignant lesion nor a precursor of carcinoma.

As used herein, the term "subject" refers to a mammalian male, preferably a human male. The subject can belong to any race or population, but is pref- erably a Caucasian male. In some embodiments, the subject can be one who has not been previously diagnosed as having prostate cancer, and who may or may not exhibit one or more signs or symptoms of prostate cancer. In some other embodiments, the subject can be one who has been previously diagnosed with prostate cancer, or identified as a subject in need of treatment for prostate cancer. He may also have already undergone treatment for prostate cancer or for one or more signs or symptoms of prostate cancer. Herein, the terms "subject", "individual", and "patient" are interchangeable.

In some embodiments, said subject is suspected to have or be at risk of having or developing prostate cancer, or is to be analysed or determined for risk of prostate cancer, genetic predisposition or susceptibility to prostate cancer, the presence or absence of prostate cancer, or the progression of prostate cancer, or is to be diagnosed for prostate cancer.

As used herein, the term "carrier status" refers to the presence or absence of a specific allele of a gene in an individual's genome.

As used herein, the term "allele" refers to any one of a series of two or more alternative forms of a gene that occupy the same position or locus on a chromosome. Different alleles are said to be polymorphic.

As used herein, the term "major allele" or "prevalent allele" refers to the allele that is found in the majority of individuals within a given population, whereas the term "minor allele" or "uncommon allele" refers to the allele that is found in the minority of individuals within a given population. With reference to genes having an impact on disease susceptibility, minor alleles are typically alleles associated with a risk of or predisposition to said disease.

The term "heterozygous" refers to individuals with one major allele and one minor at a specific locus. The term "minor homozygous" refers to indi- viduals with two minor alleles at a specific locus, whereas the term "major homozygous" refers to individuals with two major alleles at a specific locus.

Accordingly, the term "carrier" refers to a subject who carries a specific allele in his genome, either in homozygous or heterozygous form.

As used herein, the term "single nucleotide polymorphism" (SNP) re- fers to a polymorphism that occurs at a single nucleotide position in a nucleotide sequence for which two or more alternative alleles are present in a given population. SNPs include single nucleotide insertions, deletions, and substitutions, which may, but need not, result in detectable differences in gene expression or protein function. A substitution that changes a codon coding for one amino acid to a co- don coding for a different amino acid may be referred to as a missense mutation. As is well known in the art, each SNP is identifiable by a unique SNP rs ID number designated by the National Center for Biotechnology Information (NCBI).

As used herein, the term "polymorphism" refers to variation caused by the alternative alleles among individuals in a population but which variations can also be detected by other means and at different molecular levels than by detecting the sequence variation at the gene-level. Alternative molecular levels at which polymorphisms can be evaluated or determined instead of or in addition to the analysing the corresponding gene alleles can be selected from the group consisting of, for example, the corresponding gene-expression products including mRNA and/or protein.

As used herein, the term "HOXB13" or "Homeobox B13" refers to a homeobox transcription factor gene located in a cluster of HOX genes, the cytogenetic location of which is 17q21.32 and which encodes the HOXB13 protein.

As used herein, the term "HOXB13 rsl38213197" refers to a polymor- phic genomic sequence GCC T[C/T] C AAA GTA ACC (SEQ ID NO: l) at NC_000017.11:g.48728343 position (GRCh38.p2), which manifests itself at the expression level as the mRNA sequence GGT TAC TTT G[G/A]A GGC (SEQ ID NO:2). The "C allele" of the C/T SNP of rsl38213197 is the major HOXB13 allele, and it may also be referred to by the terms "non-risk allele", "non-disease- associated allele", and "HOXB13 rsl38213197 (C)". Accordingly, the "T allele" of the C/T SNP of rsl38213197 is the minor HOXB13 allele, and it may be referred to by the terms "risk allele", "disease-associated allele", "HOXB13 rsl38213197 (T)", "HOXB13 T allele", and the like.

The change of cytosine for thymine at NC_000017.11:g.48728343 position (GRCh38.p2) in HOXB13 results in amino acid variation denoted as "G84E", wherein position 84 encodes a glutamic acid (E) instead of a glycine (G). Accordingly, the G84E variant of HOXB13 may be referred to by the terms "risk variant" and "disease-associated variant". As used herein, the terms "HOXB13 G84E", "HOXB13 rsl38213197 (T)", and "HOXB13 T allele" are interchangeable.

As used herein, the term "CIP2A" refers to a gene that encodes the hu- man oncoprotein CIP2A, also known as KIAA1524 or p90, the cytogenetic location of which is 3ql3.13.

As used herein, the term "CIP2A rs2278911" refers to a polymorphic genomic sequence GTT T[C/T] G AGC ATG GAA (SEQ ID NO:3) at NC_000003.12:g.l08579413 position (GRCh38.p2), which manifests itself at the expression level as the mRNA sequence TTC CAT GCT C [G/A]A AAC (SEQ ID N0:4). The "C allele" of the C/T SNP of rs2278911 is the major CIP2A allele, and it may also be referred to by the terms "non-risk allele", "non-disease-associated allele", and "CIP2A rs2278911 (C)". Accordingly, the "T allele" of the C/T SNP of rs2278911 is the minor CIP2A allele, and it may be referred to by the terms "risk allele", "disease-associated allele", "CIP2A rs2278911 (T)", and the like.

The change of cytosine for thymine at NC_000017.11:g.48728343 position (GRCh38.p2) in CIP2A results in amino acid variation denoted "R229Q", wherein position 229 encodes a glutamine (Q) instead of an arginine (R). Accordingly, the R229Q variant of CIP2A may be referred to by the term "risk variant" or "disease-associated variant". As used herein, the terms "CIP2A R229Q", "CIP2A rs2278911 (T)", and "CIP2A T allele" are interchangeable.

As used herein, the term "HOXB13 T and CIP2A T dual carrier", and any equivalent expressions, refer to a subject who carries T alleles of both HOXB13 and CIP2A. Independently from each other, said T alleles may be present in either homozygous or heterozygous form. Thus, said dual carrier may be a subject having a genotype of HOXB13 TT and CIP2A TT, HOXB13 CT and CIP2A TT, HOXB13 TT and CIP2A CT, or HOXB13 CT and CIP2A CT. The expressions "HOXB13 T and CIP2A T", "HOXB13 T allele and CIP2A T allele", "HOXB13 rsl38213197 (T) and "CIP2A rs2278911 (T)", "HOXB13 G84E and CIP2A R229Q", and any linguistic variations thereof, are interchangeable and they all refer to the concurrent presence of T allele of HOXB13 and T allele of CIP2A.

Accordingly, the term "HOXB13 T carrier" refers to a subject who carries at least one HOXB13 T allele and who may carry either CIP2A C or T allele. In other words, said HOXB13 T carrier may be a subject having a genotype of HOXB13 TT and CIP2A TT, HOXB13 TT and CIP2A CT, HOXB13 TT and CIP2A CC, HOXB13 CT and CIP2A TT, HOXB13 CT and CIP2A CT, or HOXB13 CT and CIP2A CC. The presence of either C or T allele of CIP2A may be indicated by the term "CIP2A rs2278911 (C/T)".

The term "CIP2A T carrier" refers to a subject who carries at least one CIP2A T allele and who may carry either HOXB13 C or T allele. In other words, said CIP2A T carrier may be a subject having a genotype of HOXB13 TT and CIP2A TT, HOXB13 CT and CIP2A TT, HOXB13 CC and CIP2A TT, HOXB13 TT and CIP2A CT, HOXB13 CT and CIP2A CT, or HOXB13 CC and CIP2A CT. The presence of either C or T allele of HOXB13 may be indicated by the term "rsl38213197 (C/T)".

As used herein, the term "non-dual T carrier" refers to a subject who carries neither T allele of HOXB13 nor T allele of CIP2A T; or who carries either T allele of HOXB13 or T allele of CIP2A, but not both. In other words, a non-dual T carrier may have a genotype HOXB13 CC and CIP2A CC; genotype HOXB13 CT and CIP2A CC; genotype HOXB13 TT and CIP2A CC; genotype HOXB13 CC and CIP2A CT; or genotype HOXB13 CC and CIP2A TT.

The expression "the presence of" a specific allele or variant indicates that a subject carrying said allele or variant it at least heterozygous for the same. Thus, the expression "the presence of HOXB13 rsl38213197 (T)", for instance, is interchangeable with the expression "the presence of at least one T allele of "HOXB13 rsl38213197". Both expressions mean that the subject in question may be either heterozygous or homozygous for HOXB13 rsl38213197 (T). Corresponding nomenclature applies for CIP2A.

A subject's genetic predisposition to, susceptibility to, or risk of prostate cancer may be expressed as a probability, such as an odds ratio or hazard ratio. As used herein, the term "odds ratio" (OR) refers to the ratio of the frequen- cy of the disease in individuals having a particular marker (allele or polymorphism) to the frequency of the disease in individuals without the marker (allele or polymorphism). Odds ratios are most commonly used in case-control studies but may also be applied in cross-sectional and cohort study designs as is well known in the art.

As used herein, the term "hazard ratio" (HR) refers to the ratio of an instantaneous risk in two experimental arms over the study time period. It represents point estimate at any given point of time, thus not being a cumulative estimate like the odds ratio. Hazard ratios are usually applied in survival analysis as is well known in the art.

Thus, with respect to a disease risk, if the risk estimate (OR or HR) for a particular allele at a SNP position is greater than one, this indicates that an individual with this particular allele has a higher risk for the disease than an individual who has the other allele at the SNP position. In contrast, if the risk estimate (OR or HR) for a particular allele is less than one, this indicates that an individual with this particular allele has a reduced risk for the disease compared with an individual who has the other allele at the SNP position.

As used herein, the term "95% confidence interval" (95% CI) refers to an estimated range of values that one can be 95% certain contains the true mean of the population, the estimated range being calculated from a given set of sample data. Generally, the 95% CI is used to estimate the precision of the OR or HR. A wide CI indicates a low level of precision of the OR or HR, whereas a narrow CI indicates a higher precision of the OR or HR.

As used herein, the terms "increased risk" and "higher risk" are interchangeable and they refer to a probability that is at least about 1.5 times, at least about 1.75, at least about 2 times, at least about 3 times, at least about 4 times, at least about 6 times, at least about 7 times, at least about 8 times, at least about 9 times, at least about 10 times, at least about 15 times, at least about 20 times, at least about 25 times, at least about 30 times, at least about 35 times, or at least about 40 times higher than a corresponding probability in a relevant control. In some embodiments, said relevant control is a subject who does not carry HOXB13 rsl38213197 (T), while in some other embodiments, said relevant control is a non-dual T carrier, such as a subject who carries neither HOXB13 rsl38213197 (T) nor CIP2A rs2278911 (T). Those skilled in the art can easily choose the relevant control for each case.

Turning now to the experimental part, the present allele-level studies show that HOXB13 rsl38213197 (T) was associated with significantly increased overall risk of prostate cancer in unselected non-familial cases (OR 8.0; 95% CI 5.0-12.8; p=7.69E ²⁵ ). The association was stronger with clinically detected prostate cancer vs. controls (OR 8.6; 95% CI 5.4-13.8; p=3.1xl0 ¹⁹ ) than with screen- detected prostate cancer vs. controls (OR 5.2; 95% CI 2.8-9.9; p=3.4xlO ^-7 ). This indicates that T carriers of HOXB13 rsl38213197 have higher risk for clinically relevant prostate cancer than for clinically non-relevant cancer. Moreover, T carriers of HOXB13 rsl38213197 showed earlier than an average onset of prostate cancer among Caucasian men (HR 1.9; 95% CI, 1.6-2.2; p=6.07E-16), and significantly increased risk of progression into CRPC as compared with non-T-carriers (OR 10.8; 95% CI 5.8-19.9; p=3.1xl0 ¹⁴ ). Taken together, these results warrant the use of HOXB13 rsl38213197 (T) as a predictive, diagnostic, and/or prognostic marker of prostate cancer, especially clinically relevant prostate cancer including aggressive prostate cancer and particularly CRPC.

As also disclosed in the experimental part, the risk of prostate cancer was especially high in HOXB13 rsl38213197 T carriers with prior BPH diagnosis (OR 12.6; 95% CI, 2.8-56.8; p=0.001). This is an important result because, generally, BPH is not considered as a precancerous condition. Accordingly, in some embodiments it is advantageous to employ HOXB13 rsl38213197 (T) for determining the risk of prostate cancer, diagnosing prostate cancer and/or prognosing prostate cancer in a subject having earlier diagnosis of BPH.

The experimental part also discloses a statistically significant association between the presence of HOXB13 rsl38213197 (T) and a high PSA level (>20 ng/mL) at diagnosis (OR 1.5; 95% CI, 1.1-2.3; p=0.012) . Accordingly, in some embodiments, it is preferable to use HOXB13 rsl38213197 (T) determinations in combination with PSA measurements in determining a subject's risk of prostate cancer, diagnosing prostate cancer and/or prognosing prostate cancer. Concurrent presence of HOXB13 rsl38213197 (T) and high PSA level (>20 ng/mL) indicate the presence or risk of having or developing clinically relevant prostate cancer.

Furthermore, the experimental part shows that dual carriers of HOXB13 rsl38213197 (T) and CIP2A rs2278911 (T) have remarkably higher odds of developing prostate cancer than non-dual T carriers (OR 21.1; 95% CI 5.2-87.5; p=0.000024). Again, the association was stronger with clinically detected prostate cancer vs. controls (OR, 23.4; 95% CI 5.7-96.8, p=0.000013) than with screen- detected prostate cancer vs. control (OR 10.7; 95% CI 2.0- 58.8; p=0.006). This indicates that dual T carriers are more likely to have or develop clinically relevant prostate cancer than clinically non-relevant cancer. These results warrant the use ΟΪ Η0ΧΒ13 rsl38213197 (T) and CIP2A rs2278911 (T) as a predictive, diagnostic, or prognostic markers of prostate cancer, especially clinically relevant prostate cancer.

Moreover, dual T carriers of HOXB13 rsl38213197 and CIP2A rs2278911 showed earlier than an average onset of prostate cancer among Caucasian men (HR 2.1; 95% CI, 1.6-2.8; p=4.52E-7), warranting closer monitoring of dual carriers for the development or onset of prostate cancer.

The experimental part also demonstrates that dual T carriers of HOXB13 rsl38213197 and CIP2A rs2278911 have higher odds of developing prostate cancer with a high Gleason score (≥ 8) than non-dual T carriers (OR 2.3; 95% CI 1.1-4.8; p=0.025). Moreover, dual T carriers of HOXB13 rsl38213197 and CIP2A rs2278911 have 36.6 times higher risk of developing castration-resistant prostate cancer as compared with non-dual T carriers (OR 36.6; 95% CI 7.5- 168.5; p=0.000007). Both above results demonstrate that dual T carriers of HOXB13 rsl38213197 and CIP2A rs2278911 have higher odds of developing aggressive prostate cancer as compared to non-dual T carriers. Naturally, having aggressive prostate cancer predicts poorer prognosis. Indeed, the experimental part shows that dual carriers of HOXB13 and CIP2A T alleles have significantly worse survival rate and die earlier of prostate cancer than non-dual T carriers (HR 3.9; 95% CI 91.6-94.5; p=0.048). These results further attest the value of HOXB13 rsl38213197 (T) and CIP2A rs2278911 (T) as a predictive, diagnostic, or prognostic markers of prostate cancer, especially aggressive prostate cancer.

The herein-disclosed risk effects of HOXB13 rsl38213197 (T), particularly in combination with CIP2A rs2278911 (T), in various aspects of clinically relevant prostate cancer are remarkably higher than those of known prostate cancer-associated SNPs. Generally, the ORs of the known SNPs range between 1.04 and 2.9.

In one aspect, the present invention provides a method which comprises providing a sample obtained from a subject; and testing said sample for HOXB13 rsl38213197 and CIP2A rs2278911. More specifically, said method may be a method of analysing a subject for prostate cancer-associated markers, a clinical risk determination method, or a method of predicting prostate cancer, determining risk of prostate cancer (including determining risk of having or developing prostate cancer), determining genetic predisposition or susceptibility to pros- tate cancer, determining the presence of prostate cancer, diagnosing prostate cancer and/or prognosing prostate cancer. In other words, said method may, in some embodiments, be phrased as a method of predicting, diagnosing and/or prognosing prostate cancer. In the present method, and in any embodiments thereof, the presence of HOXB13 rsl38213197 (T), preferably in combination with CIP2A rs2278911 (T), indicates that said subject is at increased risk of hav- ing or developing prostate cancer. On the other hand, the absence of HOXB13 rsl38213197 (T) either alone or in combination with the absence of CIP2A rs2278911 (T) indicates that said subject is not at increased risk of having or developing prostate cancer.

In some embodiments of the present method, the presence of HOXB13 rsl38213197 (T), irrespective of the subject's CIP2A carrier status, i.e. the presence of CIP2A rs2278911 (C/T), indicates increased risk of prostate cancer, presence of clinically-relevant prostate cancer, presence of aggressive prostate cancer, especially CRPC, or increased risk of progression into clinically relevant or aggressive prostate cancer, especially CRPC. In some further embodiments said in- creased risk indicates earlier than an average onset of prostate cancer. In some still further embodiments, the subject to be analysed for the risk of prostate cancer, diagnosed for the presence of prostate cancer, or analysed for the risk of progression of prostate cancer, has an earlier diagnosis of BPH. In one embodiment of the present method, the subject to be analysed for the risk of prostate cancer, diagnosed for the presence of prostate cancer, or analysed for the risk of progression of prostate cancer, belongs to a family with a high incidence of prostate cancer.

In some other embodiments of the present method, the presence of HOXB13 rsl38213197 ΓΤ) and of CIP2A rs2278911 ΓΤ) indicates the presence of, a remarkably increased risk of having or developing, or a remarkable susceptibility or genetic predisposition to prostate cancer, especially clinically relevant prostate cancer, aggressive prostate cancer, or CRPC; or increased risk of progression into aggressive prostate cancer, especially CRPC. In some further embodiments said increased risk indicates earlier than an average onset of prostate cancer.

In some further embodiments, the herein-disclosed risk effects are higher in subjects who are homozygous for one or both of the present risk alleles than in subjects who are heterozygous for both of the risk alleles. In other words, the risk effects increase with the increasing number of disease-associated T alleles. A subject who is homozygous for both HOXB13 rsl38213197 (T) and CIP2A rs2278911 (T) has four (i.e. the maximum number) disease-associated T alleles. In some still other embodiments, determining susceptibility or predisposition of a subject to prostate cancer means determining the risk of said subject of getting prostate cancer prior to any diagnosis of prostate cancer. Thus, in some embodiments, the present method is to be carried out prior to any prostate can- cer diagnosis. In such embodiments, the presence of at least one T allele of the HOXB13 rsl38213197, especially together with the presence of at least one T allele of the CIP2A rs2278911, is indicative of increased risk of said subject of getting prostate cancer. Accordingly, detecting the presence of HOXB13 rsl38213197 (T), and especially detecting the concurrent presence of HOXB13 rsl38213197 (T) and of CIP2A rs2278911 (T), would justify scheduled medical examinations or more frequent screening in subjects in whom prostate cancer is not yet diagnosed, but in whom an increased risk of prostate cancer, especially aggressive prostate cancer, is identified by any one of methods disclosed herein.

Thus, the present method may be used for categorizing subjects into a high-risk group or a low-risk group on the basis of their HOXB13 and CIP2A genotype, such that subjects who are carriers of HOXB13 T allele, and especially dual carriers of HOXB13 and CIP2A T alleles, are assigned to the high-risk group, while non-carriers of HOXB13 T allele, and especially non-dual carriers of HOXB13 and CIP2A T alleles, are assigned to the low-risk group. In some embodiments, such grouping may be used for stratifying high-risk subjects not diagnosed with prostate cancer for active monitoring for the onset of prostate cancer, for early intervention, for preventive therapies, or for therapeutic trials for the same with the aim of preventing or delaying the appearance of prostate cancer or aggressive prostate cancer.

Accordingly, the present method and herein-disclosed prostate cancer risk determinations may be incorporated into a population screening protocol to identify subjects predisposed or susceptible to prostate cancer prior to onset of the disease and/or prior to increase in the PSA levels. This would enable active surveillance for the onset of prostate cancer, e.g. by more frequent PSA testing, biparametric magnetic resonance imaging (MRI), and/or digital rectal examination (DRE). Such screening would also enable the subjects to take appropriate steps aiming to prevent, delay, or reduce the severity of prostate cancer. Such appropriate steps are well known in the art and include choosing a healthy diet, reducing weight, and exercising. Moreover, early detection of prostate cancer would allow treating the disease when chances of cure are the highest. In other embodiments, determining the susceptibility, predisposition, or risk of a subject to aggressive prostate cancer is performed either prior to, at the time, or after the diagnosis of prostate cancer, wherein the presence of at least one T allele of the HOXB13 rsl38213197, especially in combination with the pres- ence of at least one T allele of the CIP2A rs2278911, is indicative of an increased risk of said subject of having aggressive prostate cancer or having prostate cancer later progressing to aggressive prostate cancer.

In some further embodiments, testing for HOXB13 rsl38213197 and CIP2A rs2278911 is carried out in a sample obtained from a subject already diag- nosed with prostate cancer. In such embodiments, the presence of HOXB13 rsl38213197 (T), and especially the simultaneous presence of HOXB13 rsl38213197 (T) and CIP2A rs2278911 (T), indicates an increased risk for the presence of, or progression to an aggressive form of prostate cancer.

The present risk determinations warrant active surveillance of high- risk subjects for the progression of prostate cancer, and provide substantial help in clinical decision making in choosing appropriate treatment procedures and in focusing resources upon those groups of subjects at highest need. For example, concurrent presence of HOXB13 and CIP2A T alleles in a subject having prostate cancer diagnosis indicates that it might be beneficial to use a more aggressive treatment strategy, such as radiotherapy or prostatectomy, instead of a less aggressive treatment strategy. On the other hand, the absence of HOXB13 rsl38213197 (T), either alone or in combination with the absence of CIP2A rs2278911 (T), indicates that less aggressive treatment would be sufficient.

Thus, the above-disclosed risk categorization of subjects on the basis of their HOXB13 and CIP2A genotype may be used for stratifying subjects into a high-risk group with aggressive prostate cancer or into a low-risk group with indolent or clinically insignificant prostate cancer. As used herein, the terms "indolent prostate cancer" and "clinically insignificant prostate cancer" are interchangeable and mean a prostate cancer with limited clinical significance, includ- ing a localized and/or slowly growing prostate cancer, a prostate cancer that is unlikely to progress to an advanced or lethal disease if left untreated, and/or a prostate cancer not related to significant morbidity or mortality of the disease. Synonyms for clinically insignificant prostate cancer include, among others, low- grade, small-volume, low-volume, minimal, organ-confined, minute and micro- focal prostate cancer. In some embodiments, the present methods are used for the stratification of subjects diagnosed with prostate cancer in a high-risk subgroup with a high probability of having or developing aggressive prostate cancer and a low- risk subgroup with a high probability of having an indolent or clinically insignifi- cant prostate cancer. In some further embodiments, subjects stratified to the high-risk and low-risk subgroups preferentially receive differential personalized therapy for prostate cancer based on their different estimated prostate-cancer related clinical endpoints. The subjects in the high-risk group are more likely to benefit from, and thus more likely to be treated by, radical prostatectomy or radi- otherapy than subjects in the low-risk group.

In some even further embodiments, subjects with a high estimated risk of developing prostate cancer or aggressive prostate based on said subject's dual carrier status of HOXB13 and CIP2A T alleles are selected for focused clinical surveillance, which may include e.g. medical examinations performed more often, or monitoring of cancer or prostate cancer-associated biomarker(s), with the aim of closer monitoring and/or earlier therapy of prostate cancer upon appearance as compared to non-dual carriers of the HOXB13 and CIP2A T alleles.

The present invention provides advantages not only for individual patient care but also for better selection and stratification of patients for clinical trials. For example, patients with significantly increased risk of aggressive forms or prostate cancer could be selected to clinical studies with the aim of developing efficient new therapeutic tools such as new medical procedures or drugs.

Furthermore, the present invention and its various embodiments may be employed in pharmacogenomic studies, which aim at analysing and predicting a response to a prostate cancer treatment or a drug. Accordingly, testing a subject for HOXB13 rsl38213197 and CIP2A rs2278911 may be carried out in any of the following cohorts: a cohort that responds favourably to a treatment regimen, a cohort that does not respond favourable to a treatment regimen, a cohort that responds significantly to a treatment regimen, a cohort that does not respond sig- nificantly to a treatment regimen, and a cohort that responds adversely to a treatment regimen (e.g. exhibits one or more side effects). These cohorts are provided as examples only, and any other cohorts or sub-cohorts may be analysed by the present method. Thus, a subject may be genotyped or tested for HOXB13 rsl38213197 and CIP2A rs2278911 to predict whether he will respond favoura- bly or significantly to a treatment regiment, not respond favourably or significantly to a treatment regimen, or respond adversely to a treatment regimen. If a sub- ject is likely to respond positively to a prostate cancer treatment with a drug, the drug may be administered to said subject. On the other hand, if a subject is likely to respond negatively (i.e. not exhibit positive effects, or exhibit adverse side effects) to prostate cancer treatment with a drug, an alternative course of treatment may be applied.

Moreover, the present invention and its various embodiments also give grounds for targeted molecular studies aiming to solve the biological background and rationale behind the synergistic effect of the HOXB13 and CIP2A T variants.

Our own recent functional studies evaluating the biological interaction between HOXB13 and CIP2A strengthen our observations of synergism in prostate cancer. Using Chromatin ImmunoPrecipitation (ChIP) with antibodies against HOXB13, coupled to massively parallel sequencing (ChlP-seq), strong chromatin binding of HOXB13 at CIP2A locus was observed both in prostate cancer cell lines and tumor tissue samples (unpublished data). HOXB13 functionally regulates the expression of CIP2A, showed by short hairpin RNA (shRNA) -mediated knockdown of HOXB13 in prostate cancer cell line. HOXB13 knockdown leads to markedly decreased level of CIP2A, suggesting that CIP2A is a direct regulatory target of HOXB13.

In the present method, and in any embodiments thereof, results from the genetic risk determinations may be combined with other test results indicative of prostate cancer, such as results of testing for one or more prostate cancer- associated biomarkers, biparametric MRI, or DRE, which were previously, concurrently, or subsequently gathered with respect to the genetic predisposition testing. Alternatively or in addition, results from the present genetic risk determination may be analysed in combination with other risk factors or clinical variables, such as age, family history, previous prostate biopsy, previous BPH, the results of a prostate exam (e.g. by biparametric MRI and/DRE) and/or body mass index. In some further embodiments, the combination of the genetic risk determi- nation results with other test results and/or risk factors or relevant clinical variables can be indicative or probative of prostate cancer, especially aggressive prostate cancer, and the combination may be utilized as an embodiment of any one of the herein-disclosed methods.

In some preferred embodiments, the determination of HOXB13 rsl38213197 and CIP2A rs2278911 is preceded by, combined with or followed by the measurement of at least one prostate cancer-associated biomarker with the purpose of increasing the predictive, diagnostic and/or prognostic value of the present method. Non-limiting examples of prostate cancer-associated biomarkers suitable for being tested from blood include total prostate specific antigen (tPSA), free prostate specific antigen (fPSA), intact prostate specific antigen (iPSA), proP- SA, human kallikrein 2 (hK2), microseminoprotein-beta (MSMB), and macrophage inhibitory cytokine-1 (MICl). In some embodiments, a combination or ratio of prostate cancer-associated biomarkers is employed. Examples of such combinations or ratios include the ratio of iPSA and tPSA either alone or in combination with the determination of hK2, the ratio of fPSA to tPSA, and the Prostate Health Index (PHI) . Non-limiting examples of prostate cancer-associated biomarkers suitable for being tested from urine include prostate cancer antigen 3 (PCA3) and TMPRSS2-ERG gene fusion product, while ERG is a non-limiting example of prostate cancer-associated tissue markers that may be detected by immunohisto- chemistry. The present method may also be used in combination with testing for any further, e.g. future or emerging, prostate cancer-associated biomarker including, but not limited to, other genetic polymorphisms in HOBX13 and/or CIP2A, genetic polymorphisms in other genes such as PTEN, RAF, BRAF, SPOP, EZH2, and Spinkl, as well as gene regulating factors (e.g. eQTLs, transcription factors, enhancers, methylation marks). Those skilled in the art know how to measure or test for these prostate cancer-associated biomarkers, and understands that also any biologically activated or inactivated derivatives or genetic variants of any the above-mentioned biomarkers may be employed, if appropriate.

The present invention and its various embodiments provide an easily applicable, simple and rapid tool for predictive prognostics and diagnostics of prostate cancer because it targets germline mutations and can hence directly be applied on any sample type containing somatic cells of the subject, not restricting the method to samples obtained by biopsy. Suitable sample types include but are not limited to whole blood, plasma, serum, saliva, urine, tissue and/or smear samples taken e.g. by swabbing the mucous membranes such as the oral mucosa.

The presence or absence of HOXB13 and CIP2A alleles of interest, preferably the T alleles, in a sample obtained from a subject can be determined by any available or future means suitable for this purpose and at different molecular levels. The determination of the T alleles can be performed at the genetic level (i.e., gene-level) by analysing DNA, or at the expression-level by analysing RNA or pro- tein. The presence or absence of the HOXB13 and CIP2A T alleles may be determined independently from each other, i.e. either at the same or different molecu- lar levels and/or by the same or methods or techniques. Some methods may allow multiplexed simultaneous testing for HOXB13 rsl38213197 and CIP2A rs2278911. Preferably, the same method is used for analysing a sample for HOXB13 rsl38213197 and CIP2A rs2278911.

Furthermore, a subject's carrier status regarding one or both of the present risk alleles (i.e., HOXB13 T allele, CIP2A T allele, or both) may be assessed directly by determining the presence or absence of said risk alleles in a sample taken from said subject. Alternatively, the occurrence of one or both of the risk alleles may be assessed indirectly by determining the presence or absence of a corresponding non-risk allele in a sample taken from said subject. In other words, absence of a non-risk allele indicates the presence of a corresponding risk allele in homozygous form, whereas, depending on the method used, presence of a non- risk allele may indicate either absence of a corresponding risk allele or that the subject is heterozygous for the risk and non-risk alleles. Generally, determining the presence or absence of the allele or variant of interest may be referred to as "testing a sample for HOXB13 rsl38213197 and CIP2A rs2278911".

Accordingly, in some embodiments, testing for HOXB13 rsl38213197 and CIP2A rs2278911 may be carried out using a method which simply indicates whether or not the HOXB13 and/or CIP2A allele of interest occurs in a subject's genome. In other words, if the allele of interest is present, the method does not tell whether the subject is heterozygous or homozygous for the presence of said allele. In some other embodiments, testing for HOXB13 rsl38213197 and CIP2A rs2278911may be carried out by a method which is able to detect both risk and non-risk alleles. Using such a method, one can determine both the presence of a risk allele in a subject's genome and whether the subject is homozygous or heterozygous for the risk allele.

Various methods for determining the presence or absence of the HOXB13 and CIP2A alleles of interest, preferably the risk alleles, are available and familiar to those skilled in the art. Accordingly, testing for HOXB13 rsl38213197 and CIP2A rs2278911may be performed, for instance, by nucleic acid sequencing. To this end, primers flanking the SNP are selected and used to amplify the region comprising the SNP. The amplified region is then sequenced using any known sequencing technique to determine which HOXB13 and/or CIP2A SNP allele(s) is (are) present in the sample. Non-limiting examples of suitable of nucleic acid se- quencing methods for use in the present invention include Sanger sequencing, NGS-based applications, single-cell sequencing, and minisequencing. Presence or absence of a particular allele may also be detected using an allele-specific amplification reaction, wherein allele-specific primers support primer extension only if the targeted allele is present. Alternatively, testing for HOXB13 rsl38213197 and CIP2A rs2278911 may be carried out by primer exten- sion assays using primers which hybridize to the nucleotides immediately upstream of the SNP. A DNA polymerase then extends the hybridized primer by incorporating a detectable nucleotide that is complementary to the SNP. Typically, detectable nucleotides comprise a label capable of producing, either directly or indirectly, a detectable signal. Similarly, any available or future nucleic acid ampli- fication assays, including but not limited to polymerase chain reaction (PCR), lig- ase chain reaction (LCR), loop-mediated amplification (LAMP), helicase- dependent amplification (HDA), nicking enzyme amplification reaction (NEAR), transcription-mediated amplification (TMA), nucleic acid sequence based amplification (NASBA), strand displacement amplification (SDA) and various other tem- perature-dependent or isothermal amplification methods may be employed in testing a sample for HOXB13 rsl38213197 and CIP2A rs2278911.

In further embodiments, determining the presence or absence of the HOXB13 and CIP2A alleles of interest may involve differential susceptibility of the polymorphic site to restriction enzymes. For this purpose, PCR may be used to amplify the region comprising the SNP. The amplified PCR product is then subjected to a restriction enzyme digestion followed by electrophoretical analysis of the restriction product. If a specific SNP allele is associated with a specific restriction site, digestion of that site is indicative of the presence of that particular SNP.

Many further types of allelic discrimination assays are known in the art. A subject's HOXB13 rsl38213197 and/or CIP2A rs2278911 genotype may also be determined using an oligonucleotide probe specific to either C or T allele of HOXB13 or CIP2A. In some embodiments, a region comprising the SNP is amplified prior to, or concurrent with, carrying out a hybridization reaction with the probe. Hybridization of the probe to the amplification product under stringent hybridization conditions indicates that the amplified region contains that particular allele of HOXB13 or CIP2A for which the probe is specific for.

Further methods suitable for determining the presence or absence of HOXB13 and CIP2A alleles of interest at the nucleic acid or genetic level include, but are not limited to, RNase protection assays, molecular beacon-based oligonucleotide binding assays, melting curve analysis combined with oligonucleotide probes and/or intercalating labels, gel electrophoresis analysis, Southern blotting, microarrays such as DNA microarrays and RNA microarrays, direct probing, and signal accumulation assays with or without amplification of the target nucleic acids.

As set forth above, determining a subject's carrier status for HOXB13 rsl38213197 and/or CIP2A rs2278911, i.e. determining the presence or absence of HOXB13 and CIP2A alleles of interest, may be carried out at protein level. The presence of HOXB G84E or CIP2A R229Q is a direct indication of the presence of the corresponding risk alleles (T alleles). Non-limiting examples of methods suit- able for protein-level analysis include mass spectrometry, immunoassays, im- munohistochemistry, and other protein-binding assays. In some specific embodiments, one or more antibodies or antibody fragments specific for a protein or polypeptide encoded by a particular polymorphic form of HOXB13 or CIP2A may be used for the determination. Binding bodies alternative to antibodies include but are not limited to aptamers, receptors and biologically interacting proteins, such as up- or down regulating targets identified in relevant biological signalling pathways.

Any available or future mRNA detection technique may be applied for determining the presence or absence of HOXB13 and CIP2A alleles of interest at the expression level. Non-limiting examples of methods suitable for mRNA-level analysis include reverse transcriptase PCR (RT-PCR), RNA microarrays, and mass spectrometry.

The present invention also provides a kit for use in the present method and any embodiments thereof. The kit comprises a first binding body for testing said sample for HOXB13 rsl38213197 and a second binding body for testing said sample for CIP2A rs2278911. Said first binding body may be specific for determining the presence or absence of either HOXB13 rsl38213197 (T) or HOXB13 rsl38213197 (C). Independently from the specificity of the first binding body, said second binding body may be specific for determining the presence or ab- sence of either CIP2A rs2278911 (T) or CIP2A rs2278911 (C). In some embodiments, the kit may even comprise a first binding body for specifically determining the presence or absence of HOXB13 rsl38213197 (T) and a different first binding body for specifically determining the presence or absence of HOXB13 rsl38213197 (C) and/or a second binding body for specifically determining the presence or absence of CIP2A rs2278911 (T) and a different second binding body for specifically determining the presence or absence of CIP2A rs2278911 (C). Such embodiments are typically employed for determining whether the sample to be analyzed with the kit is homozygous or heterozygous for HOXB13 rsl38213197 and/or CIP2A rs2278911.

Depending on the detection technique and the level of detection (i.e. detection at gene-level, mRNA-level, or protein-level) to be employed, those skilled in the art can easily choose appropriate first and second binding bodies to be included in the kit, for determining the presence or absence of HOXB13 rsl38213197 (T), HOXB13 rsl38213197 (C), CIP2A rs2278911 (T) and/or CIP2A rs2278911 (C). For example, said first and second binding bodies may be selected, independently from each other, from the group consisting of nucleic acid primers, nucleic acid probes, aptamers, antibodies or antibody fragments, peptides, receptors, and enzymes, including restriction enzymes. Those skilled in the art can readily choose appropriate binding bodies depending on the detection technique to be employed.

For detection, binding bodies may be conjugated or otherwise associated with a detectable label selected from the group including, but not limited to, optical agents such as luminescent or fluorescent labels including a variety of organic and/or inorganic small molecules and a variety of fluorescent proteins and derivatives thereof, phosphorescent labels, chemiluminescent labels, and chro- mogenic labels; radioactive labels such as radionuclides that emit gamma rays, positrons, beta or alpha particles, or X-rays; and enzymes such as alkaline phosphatase (AP), luciferase, or (horseradish) hydrogen peroxidase (HRP). Said association can be direct, e.g. through a covalent bond, or indirect, e.g. via another binding agent, chelator, or linker. Also, the label may be directly detectable or in- directly detectable through a detectably labelled secondary binding body. Detectable labels and techniques for conjugating or otherwise associating said detectable labels to binding bodies are well known in the art. Furthermore, those skilled in the art can readily choose appropriate detectable labels depending on the type of binding bodies to be employed.

It will be obvious to a person skilled in the art that, as technology advances, the inventive concept can be implemented in various ways. The invention and its embodiments are not limited to the examples described below but may vary within the scope of the claims. EXAMPLES

Example 1. Sequencing and SNP genotyping of study subjects

Study subjects

The study was performed in Finland and all cancer cases and controls involved and genotyped in this study were of Caucasian origin. Written informed consent was obtained from each study subject. The study protocol was approved by the research ethics committee at Pirkanmaa Hospital District (Tampere, Finland) and by the National Supervisory Authority for Welfare and Health (VALVI- RA).

For HOXB13 mutation 2669 and for CIP2A variant 2738 unselected non-familial eligible prostate cancer (PrCa) cases were analyzed, respectively. The genotyping was partly carried out by the PRACTICAL (PRostate cancer Association group To Investigate Cancer Associated alterations in the genome) consortium. Of cases, 2281 were clinically diagnosed cases from the Pirkanmaa Hos- pital District, confirmed from medical records. Another set of subjects consisted of 457 Finnish screen detected cancer cases recruited by the Finnish arm of The European Randomized Study of Screening for Prostate Cancer (ERSPC). This study was initiated in the mid-1990s to evaluate the effect of prostate-specific antigen (PSA) screening on death rates from PrCa (Schroder et ah, N Engl J Med. 2009;360: 1320-8). Of the patients, 106 were young-onset cases.

PrCa control subjects [HOXB13 G84E, n=2423; CIP2A R229Q, n=2427), belonging to the screening trial control group, were derived from the Finnish arm of the ERSPC (Schroder et ah, Ibid). Control subjects were population-matched healthy individuals of ages between 70 and 86 years who had undergone PSA screening. Their disease status is annually evaluated from the records of the Finnish Cancer Registry.

Clinical and pathological sub-classification of PrCa patients

Clinically detected PrCa cases (83.3%) dominated the sample set compared to a smaller proportion of screen-detected patients (16.7%) . Notably, 27.4% of all cases developed either PSA progression, local or distant progression, and 10.5% emerged castration resistance after androgen deprivation therapy. From 1126 deceased PrCa patients 26.4% died of PrCa, which is 10.9% of the whole PrCa population studied. In addition, germline DNA was available from altogether 590 clinically and pathologically defined BPH from the Urology Outpa- tient Clinic in Tampere University Hospital (Tampere, Finland). From these, 198 (33.6 % of BPH patients) developed PrCa later in life (7.23% of all PrCa cases). Clinical characteristics of genotyped PrCa patients are summarized in Table 1.

Table 1. Clinical characteristics of prostate cancer patients

Missing data 556 (20.3)

SNP genotyping and sequencing

Germline blood DNA was collected and isolated according to standard methods, and genotyped for HOXB13 G84E (rsl38213197) using a Custom Taq- Man SNP assay (Applied Biosystems/Life Technologies) and for CIP2A R229Q (rs2278911) using custom Illumina iSelect SNP genotyping array platform, designed as part of the Collaborative Oncological Gene-Environment Study (COGS), according to the manufacturer's instructions. BPH samples for HOXB13 G84E were genotyped by the Technology Centre, Institute for Molecular Medicine Fin- land (FIMM), University of Helsinki (Helsinki, Finland) using the MassARRAY iPLEX platform (Sequenom, Inc.). Duplicate test samples and four negative controls were included in each 384-well plate. Concordance for the control samples was >99%. In selected set of PrCa samples the studied variants were confirmed through standard Sanger sequencing using ABI PRISM BigDye Termination Cycle Sequencing Ready Reaction Kit (Applied Biosystems/Life Technologies).

Example 2. Risk of prostate cancer

The Hardy-Weinberg equilibrium equation was used to determine whether the proportion of each genotype obtained was in agreement with the expected values as calculated from the allele frequencies.

Statistical analyses were performed with IBM SPSS version 22 (SPSS

Inc, Chicago, USA) unless otherwise specified. As the number of parameters was relatively small compared to the number of subjects unconditional logistic regression analyses was used to measure the association between HOXB13 and CIP2A variants and the risk of PrCa by estimation of odds ratio (OR) and its 95% confi- dence interval (CI) . P-values were 2-sided and p<0.05 was considered to indicate a statistically significant result.

According to the results, the HOXB13 G84E and CIP2A R229Q mutations were in Hardy-Weinberg equilibrium in both cases (p=0.097 and p=0.739) and controls (p=0.838 and p=0.498), respectively. The overall minor allele fre- quency for G84E variant was 1.8%, and for R229Q was 13.8% in the entire sample set. The G84E mutation was detected in 187 subjects, of which 167 were patients with prostate cancer (carrier frequency 6.24%) and 20 were healthy controls (carrier frequency 0.80%) . The carrier frequency of R229Q variant was 24.7% (n=675) in cases and 26.5% (n=642) in controls (Table 2.).

H0XB13 G84E T allele was associated with 8.0-fold increased risk of PrCa (95% CI 5.0-12.8, p=7.69E-25) in unselected non-familial cases in analyzed Finnish samples.

Significant result was seen for H0XB13 T and CIP2A T dual carriers as they showed 21.1 times increased risk for PrCa (95% CI 5.2-87.5; p=0.000024).

Example 3. Dissection of the effect of HOXB 13 G84E and CIP2A R229Q vari- ants on prostate cancer risk

To evaluate the relative effects of H0XB13 and CIP2A variants on PrCa development binary stepwise logistic-regression with backward elimination method was used as described by Cordell et al. (Am J Hum Genet. 2002; 70: 124-41.37). By fitting statistical models with main effects, a test with few degrees of freedom that is likely to be powerful for detecting primary etiological determinants was employed. This approach is applicable to case-control studies modelled via unconditional logistic regression. By testing two polymorphisms, a small number of degrees of freedom will be present.

In other words, the effects of HOXB13 and CIP2A variants on PrCa risk were tested with three selection methods (enter, forward and backward) of binary stepwise logistic-regression. The usually preferred backward elimination method gave the most significant result. This procedure was started with all candidate variants, and tested if the deletion of the variant improves the model the most by being deleted, and repeating this process until no further improvement is possible. All of the genetic variants, HOXB13 T carrier, CIP2A T carrier and dual minor allele carrier status (TT), were included in the initial model and a backward stepwise regression selection procedure was used, with an exit p-value of 0.05, to create the final model. Removing HOXB13 T carriers deteriorated the model most, change in -2 log likelihood was 76.365 points, the significance of change was p=1.3681E-28. Additionally, removing of dual T carriers deteriorated the model as well, but with lower effect (change in -2 Log Likelihood is 4.330 points, P=0.037).

When using all the predictor genetic variants as a group (ENTER method) the Nagelkerke R Square test revealed that 3.3% of the variability in PrCa risk can be accounted for by all of the used predictors together in the Finnish patients. With this method the HOXB13 T carriers showed significant risk for PrCa (OR 6.268; 95% CI, 3.804-10.327; p=5.8567E-13); but none of other tested variables.

In this way, regression analysis confirmed that HOXB13 variant has the main effect in PrCa development of all variants used in analyses model.

Example 4. Time to prostate cancer development

Time to event analyses (carriers vs. non-carriers) were carried out by Cox regression method. Results are shown in Table 3 below.

The results showed significant risk for earlier PrCa development in H0XB13 T allele carriers (HR 1.9; 95% CI, 1.6-2.2; p=6.07E-16), whereas a synergistic effect in dual carriers of H0XB13 T and CIP2A T alleles was shown not only in increasing susceptibility, but also in earlier development of the disease (HR 2.1; 95% CI, 1.6-2.8; p=4.52E-7).

Example 5. Correlation with clinical features

Binary logistic regression analyses were conducted to evaluate the impact of the mutations on tumour phenotype and selected clinical features, including PSA at diagnosis, Gleason-score, general progression, clinical and screen detection, CRPC development, BPH status and PrCa development, general and disease specific death.

Statistically significant associations of HOXB13 G84E carriers, and the dual carriers of HOXB13 G84E and CIP2A R229Q (vs. non-carriers) with clinical characteristics are presented in Table 4. No evidence of association of CIP2A R229Q carrier status and clinical features of examined PrCa patients was found.

Significant association was observed between the HOXB13 mutation status and high PSA at diagnosis (PSA>20 ng/mL; OR 1.5; 95% CI 1.1-2.3; p=0.012); total PSA data were available for 94.3% of cases. No significant association was observed between the mutation status and Gleason score presented (p=0.093).

Clinical features of dual HOXB13 G84E and CIP2A R229Q carriers were examined compared with non-dual T carriers. Significant association was found between dual T carriers of HOXB13 and CIP2A variants and a high Gleason score (Gleason≥ 8, OR 2.3; 95% CI 1.1-4.8; p=0.025), which is a commonly used clinical parameter defining aggressive, poorer prognosis predicting cancer.

Example 6. Association with specific prostate cancer subgroups

Applied case-control and direct subgroup comparison analyses of clinically differentiated PrCa and the HOXB13 G84E, CIP2A R229Q carrier status revealed associations with several hallmarks of the disease (Table 5.).

H0XB13 mutation showed strong association with clinically detected PrCa cases (OR 8.6; 95% CI 5.4-13.8; p=3.1x10-19) vs. controls, and lower association with screen-detected PrCa vs. controls (OR 5.2; 95% CI 2.8-9.9 p=3.4x10-7). HOXB13 T allele's role in clinically more severe disease was confirmed when clini- cally detected cases vs. screen detected PrCa were compared (OR 1.6; 95% CI 1.0- 2.7 p=0.046). Although BPH is not considered as premalignant lesion, the present study showed that men with G84E variant with prior BPH diagnosis have 12.6- fold risk to develop PrCa compared to BPH controls (BPH, but no later development of PrCa; 95% CI 2.8-56.8; p=0.001). HOXB13 T carrier status conferred for increased risk of PrCa in men with prior BPH diagnosis compared to men with BPH but with no later development of PrCa (OR 12.6; 95% CI 2.8-56.8 p=0.001). Importantly, statistically significant evidence of HOXB13 T allele's contribution to CRPC risk compared to controls was detected (OR 10.8; 95% CI 5.8-19.9; p=3.1x10-14).

Dual T carriers of HOXB13 and CIP2A had higher risk of clinically detected prostate cancers compared with controls (OR, 23.4; 95% CI 5.7-96.8, p=0.000013) and a lower chance to be recognized through PrCa screening vs. controls (OR 10.7; 95% CI 2.0- 58.8; p=0.006) . The risk for CRPC vs. controls was more than three times higher (OR 36.6 (95% CI 7.5-168.5), p=0.000007) when the patient was dual T carrier of HOXB13 and CIP2A compared to patients having only the HOXB13 T allele.

Even though the association of dual T carriers and PrCa development in BPH diagnosed men could not be statistically evaluated because of the small sample size, it is noteworthy to mention that from 590 BPH samples two were dual T carriers and they both had PrCa diagnosis. Especially important was to evaluate the risk of studied variants on development of castration resistance during the treatment, i.e., the risk of the subjects to develop castration-resistant prostate cancer.

After thoroughly examined treatment modalities, we found that pa- tients treated with prostatectomy had a higher mutation carrier rate for HOXB13 T, CIP2A T, and dual T carriers (7.4%, 25.2%, 2.2%, respectively) compared to patients on active surveillance (5.5%, 21.7%, 1.6%, respectively). Carriers of the HOXB13 and/or CIP2A T alleles were hence more likely to receive primary oncological treatment. Example 7. Survival

In order to explore the role of H0XB13 and CIP2A in overall and in PrCa specific survival, Kaplan-Meier survival analyses were applied. Survival time (years) was compared between H0XB13 and CIP2A T carriers and those without the mutations.

Exploring the longest follow up time (birth-death), dual T carriers of HOXB13 and CIP2A had significantly worse survival and they died earlier of PrCa (Breslow test, HR 3.9; 95% CI 91.6-94.5; p=0.048) vs. non dual T carriers.

The combination of HOXB13 T and CIP2A T alleles seemed to have def- inite effect on PrCa risk, time to develop the disease, aggressive disease, survival and PrCa specific death.

Example 8. In silico pathogenicity prediction

To further explore the function of HOXB13 G84E and CIP2A R229Q variants in PrCa, and to investigate the pathogenicity prediction of G84E and R229Q variants, an in silico tool for missense variants, named Combined Annotation Dependent Depletion (CADD) prediction software, was used. This enables the objective integration of many diverse annotations into a single measure (C-score) for each variant. This software integrates diverse genome annotations, using altogether ~63 different tools, applying conservation matrices as well as protein based matrices. The basis of CADD is to contrast the annotations of fixed alleles in humans with those of simulated variants. A variant is predicted to be pathogenic if the scaled C-score calculated by the software is above 10, which indicates the top 10% deleterious state among possible substitutions. For comparison the Human Genome version 37pl3 (hgl9) was used.

HOXB13 G84E was predicted to be deleterious with a scaled C-score of

22.7, which indicates a potentially causative mutation. CIP2A R229Q variant had a scaled C-score of 21.9, denoting a similarly deleterious variant as well as a mutation potentially predisposing to cancer.