Prostate Cancer

A number sign (#) is used with this entry because of evidence that many genes are involved in the origin and/or progression of this neoplasm.

Genetic Heterogeneity of Susceptibility to Prostate Cancer

See HPC1 (601518), associated with variation in the RNASEL gene on chromosome 1q25; HPC2 (614731), associated with variation in the ELAC2 gene (605367) on chromosome 17p12; HPC3 (608656), mapped to chromosome 20q13; HPC4 (608658), mapped to chromosome 7p11-q21; HPC5 (609299), mapped to chromosome 3p26; HPC6 (609558), mapped to chromosome 22q12; HPC7 (610321), mapped to chromosome 15q12; HPC8 (602759), mapped to chromosome 1q42.2-q43; HPC9 (610997), associated with variation in the HOXB13 gene (604607) on chromosome 17q21-q22; HPC10 (611100), mapped to chromosome 8q24; HPC11 (611955), mapped to chromosome 17q12; HPC12 (611868), associated with variation in the EHBP1 gene (609922) on chromosome 2p15; HPC13 (611928), associated with variation in the MSMB gene (611928) on chromosome 10q11; HPC14 (611958), mapped to chromosome 11q13; HPC15 (611959), mapped to chromosome 19q13; HPCX1 (300147), mapped to chromosome Xq27-q28; and HPCX2 (300704), mapped to chromosome Xp11.

Somatic mutations in several genes have been found in prostate cancer tumors, including PTEN (601728), MAD1L1 (602686), ATBF1 (ZFHX3; 104155), and KLF6 (602053).

Somatic mutations in the EPHB2 gene (600997) are associated with susceptibility to prostate cancer/brain cancer (603688).

A prostate cancer aggressiveness quantitative trait locus (HPCQTL19; 607592) has been mapped to chromosome 19q.

Also see MOLECULAR GENETICS.

Clinical Features

The aggressiveness of prostate cancer varies widely. Some tumors progress to invasive, potentially life-threatening disease, whereas others stay latent for the remainder of an individual's lifetime.

Inheritance

Studies by Woolf (1960), Cannon et al. (1982), Meikle et al. (1985) and others suggested the significance of familial factors in prostate cancer.

Steinberg et al. (1990) investigated the frequency of prostate cancer in the relatives of 691 men with prostate cancer and 640 of their spouses. In 15% of the cases, but only 8% of the controls, it was a father or brother affected with prostate cancer (P less than 0.001). Men with a father or brother affected were twice as likely to develop prostate cancer as men with no relatives affected. There was a trend of increasing risk with increasing number of affected family members, such that men with 2 or 3 first-degree relatives affected had a 5- and 11-fold increased risk of developing prostate cancer, respectively. Carter et al. (1992) found in the 691 prostate cancer families that early onset of disease in the proband was also an important determinant of risk. Complex segregation analysis led to the conclusion that the family clustering was best explained by autosomal dominant inheritance of a rare (q = 0.0030) high-risk allele leading to an early onset of prostate cancer. The estimated cumulative risk of prostate cancer for carriers showed that the allele was highly penetrant; by age 85, 88% of carriers compared to only 5% of noncarriers were projected to be affected with prostate cancer. Twin studies support the existence of a significant genetic factor (Walsh, 1992).

To determine whether family history is associated with an increased prevalence of prostate cancer in an unselected group of men attending a hospital-based screening clinic, Narod et al. (1995) inquired about affected relatives before prostate cancer screening in 6,390 men, aged 50 to 80 years, in the region of Quebec City. Of these 6,390 men, 1,563 (24.5%) had a positive test by either rectal examination or blood test for prostate-specific antigen (APS; 176820). Among the 6,390 men, 264 were found to have prostate cancer (4.13%). The prevalence was increased in those men with any first-degree relative affected (prevalence = 6.7%; relative risk = 1.72 as compared with men with no first-degree relative affected; prevalence = 3.89; relative risk = 1.00). Most of the increase in relative risk was contributed by affected brothers (prevalence = 10.2%; relative risk = 2.62; P = 0.0002).

Schaid et al. (1998) performed a family history cancer survey on 5,486 men who underwent radical prostatectomy at the Mayo Clinic for clinically localized prostate cancer; 4,288 men responded to the survey. The best-fitting model that explained familial aggregation was a rare autosomal dominant susceptibility gene, and this model fit best when probands were diagnosed under 60 years of age. The model predicted that the frequency of a susceptibility gene in the population is 0.006 and that the risk of prostate cancer by age 85 years is 89% among carriers of the gene and 3% among noncarriers. Genetic heterogeneity was suggested by the study. Other evidence suggesting genetic complexity included the significantly elevated age-adjusted risk of prostate cancer among brothers of probands, compared with their fathers.

Cui et al. (2001) conducted single- and 2-locus segregation analyses of data from 1,476 men with prostate cancer diagnosed under the age of 70 years and ascertained through population registers in Australia, together with brothers, fathers, and both maternal and paternal lineal uncles. All 2-locus models gave better fits than did single-locus models, even if lineal uncles were excluded. The best-fitting genetic models included a dominantly inherited increased risk that was greater, in multiplicative terms, at younger ages, as well as a recessively inherited or X-linked increased risk that was greater, in multiplicative terms, at older ages. Penetrance to age 80 years was approximately 70% for the dominant effect and virtually 100% for the recessive and X-linked effects.

Nieder et al. (2003) suggested that major risk factors for developing prostate cancer, including positive family history and African American ethnicity, can be quantified for genetic counseling. Factors increasing familial risk for prostate cancer are closer degree of kinship, number of affected relatives, and early age of onset (under 50 years) among the affected relatives. Genetic testing may be useful for modification of risk, but at the time of writing, Nieder et al. (2003) suggested that genetic testing should be performed only within the context of a well-designed research study that will determine penetrance and genotype-phenotype correlation of specific mutations. Even in the absence of genetic testing, African American men and men with a strong family history of prostate cancer may opt to initiate screening by prostate-specific antigen and digital rectal examination at age 40 years.

Valeri et al. (2003) stated that recent linkage analyses had led to the detection of at least 8 prostate cancer predisposing genes, suggesting complex inheritance and genetic heterogeneity. They conducted segregation analysis in 691 prostate cancer patients, recruited from 3 French hospitals and unselected with respect to age at diagnosis, clinical stage, or family history. Segregation analyses were carried out using the logistic hazard regressive model, as incorporated in the REGRESS program, which can accommodate a major gene effect, residual familial dependences of any origin (genetic and/or environmental), and covariates, while including survival analysis concepts. Segregation analysis showed evidence for the segregation of an autosomal dominant gene (allele frequency 0.03%) with an additional brother-brother dependence. The estimated cumulative risks of prostate cancer by age 85 years, among subjects with the at-risk genotype, were 86% in the fathers' generation and 99% in the probands' generation. The study supported the model of mendelian transmission of a rare autosomal dominant gene with high penetrance, and demonstrated that additional genetic and/or common sib environmental factors are involved to account for the familial clustering of prostate cancer.

Pakkanen et al. (2007) performed segregation analysis in 2 cohorts of 557 early-onset and 989 late-onset nuclear Finnish families in which the father had histologically confirmed prostate cancer. Their findings suggested that inheritance of prostate cancer in the Finnish population is best explained by a mendelian recessive model with a significant paternal regressive coefficient that is indicative of a polygenic multifactorial component.

Mapping

To investigate genetic factors involved in the variability of prostate cancer aggressiveness, Witte et al. (2000) conducted a genomewide linkage analysis of 513 brothers with prostate cancer, using the Gleason score, which reflects tumor histology, as a quantitative measure of prostate cancer aggressiveness. To their knowledge, this was the first time that a measure of prostate cancer aggressiveness had been directly investigated as a quantitative trait in a genomewide scan. Candidate regions were found on 5q, 7q, and 19q (see 607592).

Gibbs et al. (2000) found evidence of linkage of susceptibilty to prostate cancer at multiple sites in a genomewide scan. Stratification by a variety of factors appeared to improve the chances of identifying relevant genes.

Ostrander and Stanford (2000) reviewed the search for prostate cancer genes. Peters and Ostrander (2001) presented a tabulation of the cytogenetic location of 16 mapped prostate cancer susceptibility loci and candidate genes.

Oba et al. (2001) studied the significance of the loss of heterozygosity (LOH) frequently observed on chromosome 8p in prostate cancer. By fluorescence in situ hybridization (FISH) in 42 prostate cancers, they found a deletion for at least 1 locus on 8p in 29 (69%) tumors. A significantly higher frequency of the deletion on 8p21.2-p21.1 was observed in advanced prostate cancer than in localized prostate cancer. They concluded that deletions on 8p22-p21.3 play an important role in tumor differentiation, while 8p21.2-p21.1 deletion plays a role in progression of prostate cancer.

Xu et al. (2001) undertook a systematic evaluation of linkage across chromosome 1 using 50 microsatellite markers in 159 hereditary prostate cancer families, including 79 families analyzed in the original report by Smith et al. (1996) describing HPC1 linkage. The results of the new study were consistent with the heterogeneous nature of hereditary prostate cancer and the existence of multiple loci on chromosome 1 for this disease.

Goddard et al. (2001) detected linkage near 3 locations for prostate cancer: 1q24-q25, 1q42.2-q43, and near Xq12-q13, the AR (313700) locus. Six other locations gave lod scores greater than 2.5.

Xu et al. (2001) performed linkage analysis using 24 markers from 8p in 159 pedigrees with hereditary prostate cancer. In 79 families with an average age at diagnosis of greater than 65 years, an allele-sharing lod score of 2.64 (P = 0.0005) was observed. Of note, the small number (11) of Ashkenazi Jewish pedigrees analyzed in this study contributed disproportionately to this linkage. Mutation screening in hereditary prostate cancer probands and association analyses in case subjects and unaffected control subjects was carried out for a putative prostate cancer susceptibility gene, which they called PG1, previously localized to the 8p22-p23 region. (PG1 had been cloned in a haplotype-based association study conducted by Geneset by Daniel Cohen and described only in a US patent dated August 31, 1999.) Xu et al. (2001) concluded that evaluation of PG1 and other candidate genes in the region appeared warranted.

Cancel-Tassin et al. (2001) examined evidence for linkage to the HPC1, HPC8, CAPB, and HPCX loci in 64 families from southern and western Europe with at least 3 affected individuals with prostate cancer. No significant evidence of linkage to HPC1, CAPB, or HPCX was found. Results in favor of linkage to HPC8 were observed and homogeneity analysis gave an estimated proportion of families with linkage to this locus up to 50%.

Schaid et al. (2004) compared genome linkage scans for prostate cancer susceptibility loci using microsatellites with those using SNPs, performed in 467 men with prostate cancer from 167 families. The highest lod scores were found for chromosome 6 (4.2) and chromosome 12 (3.9), but these were judged difficult to interpret because they occurred only at the extreme ends of the chromosomes. The greatest gain provided by the SNP markers was a large increase in the linkage information content, with an average information content of 61% for the SNPs versus an average of 41% for the microsatellite markers.

Paris et al. (2004) analyzed a cohort of 64 prostate cancer patients (half of whom had postoperative recurrence) using array comparative genomic hybridization (aCGH). Analysis of the aCGH profiles revealed numerous recurrent genomic copy number aberrations. Specific loss at 8p23.2 was associated with advanced stage disease, and gain at 11q13.1 was found to be predictive of postoperative recurrence independent of stage and grade. Moreover, comparison with an independent set of metastases revealed approximately 40 candidate markers associated with metastatic potential. The authors proposed that copy number aberrations at these loci may define metastatic genotypes.

Xu et al. (2005) performed genomewide linkage analysis of 269 prostate cancer families with at least 5 affected members and found significant linkage at 22q12 (lod score, 3.57; HPC6; 609558). They also found 'suggestive' linkage (lod score of 1.86 or greater) at 1q25, 8q13, 13q14, 16p13, and 17q21 in these families. In 606 families with prostate cancer with a mean age at diagnosis of 65 years or less, 4 additional suggestive linkages were found: 3p24, 5q35, 11q22, and Xq12.

Thomas et al. (2008), in a genomewide association study (GWAS) of prostate cancer, confirmed 3 previously reported loci: 2 independent SNPs at 8q24 (HPC10; 611100) and 1 in HNF1B (189907) on 17q (HPC11; 611955). In addition, loci on chromosomes 7, 10 (2 loci), and 11 (HPC14; 611958) were highly significant. Loci on chromosome 10 (see HPC13, 611928) included MSMB (157145), which encodes beta-microseminoprotein, a primary constituent of semen and a proposed prostate cancer biomarker, and CTBP2 (602619), a gene with antiapoptotic activity. The locus on chromosome 7 was at JAZF1 (606246), a transcriptional repressor that is fused by chromosome translocation to SUZ12 (606245) in endometrial cancer. Of the 9 loci that showed highly suggestive associations, 4 best fit a recessive model and included candidate susceptibility genes: CPNE3 (604207), IL16 (603035), and CDH13 (601364). The findings pointed to multiple loci with moderate affects associated with susceptibility to prostate cancer that, taken together, in the future may predict high risk in select individuals.

Eeles et al. (2008) conducted a genomewide association study using blood DNA samples from 1,854 individuals with clinically detected prostate cancer diagnosed at or before the age of 60 years or with a family history of disease, and 1,894 population-screened controls with a low prostate-specific antigen (PSA; 176820) concentration (less than 0.5 ng/ml). They analyzed these samples for 541,129 SNPs using the Illumina Infinium platform. Initial putative associations were confirmed using a further 3,268 cases and 3,366 controls. They identified 7 loci associated with prostate cancer on chromosomes 3, 6, 7, 10, 11 (HPC14), 19 (HPC15; 611959), and X. They confirmed previous reports of common loci associated with prostate cancer at 8q24 (HPC10; 611100) and 17q (HPC11; 611955). Three of the newly identified loci contained candidate susceptibility genes: MSMB (157145) on 10q11.2, LMTK2 (610989) on 7q21.3-q22.1 and KLK3 (176820) on 19q13.4.

Eeles et al. (2009) extended the study of Eeles et al. (2008) to evaluate promising associations in a second stage in which they genotyped 43,671 SNPs in 3,650 prostate cancer cases and 3,940 controls, and in a third stage involving an additional 16,229 cases and 14,821 controls from 21 studies. In addition to replicating previous associations, Eeles et al. (2009) identified 7 new prostate cancer susceptibility loci on chromosomes 2, 4, 8, 11, and 22 (with P = 1.6 x 10(-8) to P = 2.7 x 10(-33)). The strongest association was found with a G/A SNP at chromosome 11p15, rs7127900 (position 2,190,150; per-allele OR = 1.22, 95% CI = 1.17-1.27, P = 2.7 x 10(-33)).

Kote-Jarai et al. (2011) extended the multistage genomewide association study of Eeles et al. (2008, 2009) and reported the results of stage 3, in which they evaluated 1,536 SNPs in 4,574 individuals with prostate cancer (cases) and 4,164 controls. They followed up 10 new association signals through genotyping in 51,311 samples in 30 studies from the Prostate Cancer Association Group to Investigate Cancer Associated Alterations in the Genome (PRACTICAL) consortium. In addition to replicating previously reported loci, they identified 7 new prostate cancer susceptibility loci on chromosomes 2p11, 3q23, 3q26, 5p12, 6p12, 12q13, and Xq12 (p = 4.0 x 10(-8) to p = 2.7 x 10(-24)). Kote-Jarai et al. (2011) identified a SNP in TERT (187270) on chromosome 5p15 (rs2242652, p = 4.4 x 10(-11)) that was more strongly associated with prostate cancer than the previously reported rs401681 and rs2736098 (Rafnar et al., 2009). Kote-Jarai et al. (2011) concluded that more than 40 prostate cancer susceptibility loci, explaining approximately 25% of the familial risk in this disease, have been identified.

Gudmundsson et al. (2009) reported a prostate cancer genomewide association follow-up study and discovered 4 variants associated with susceptibility to prostate cancer in several European populations: rs10934853A (OR = 1.12, P = 2.9 x 10(-10)) on 3q21.3; 2 SNPs on chromosome 8q24.21, as described in HPC10 (611100); and rs8102476C (odds ratio = 1.12, P = 1.6 x 10(-11)) on 19q13.2. Gudmundsson et al. (2009) also refined a previous association signal on 11q13 (see HPC14, 611958). In a multivariate analysis using 22 prostate cancer risk variants typed in the Icelandic population, Gudmundsson et al. (2009) estimated that carriers in the top 1.3% of the risk distribution are at a 2.5 times greater risk of developing the disease than members of the general population.

Cytogenetics

Tomlins et al. (2005) used a bioinformatics approach to discover candidate oncogenic chromosomal aberrations on the basis of outlier gene expression. Two ETS transcription factors, ERG (165080) and ETV1 (600541), were identified as outliers in prostate cancer. Tomlins et al. (2005) identified recurrent gene fusions of the 5-prime untranslated region of TMPRSS2 (602060) to ERG or ETV1 in prostate cancer tissues with outlier expression. By using FISH, Tomlins et al. (2005) demonstrated that 23 of 29 prostate cancer samples harbored rearrangements in ERG or ETV1. Cell line experiments suggested that the androgen-responsive promoter elements of TMPRSS2 mediate the overexpression of ETS family members in prostate cancer.

Chang et al. (2006) performed 2-locus conditional linkage analysis to identify possible gene-gene interactions in a genomewide scan of 426 families with prostate cancer. Suggestive evidence for an epistatic interaction was found for 6 sets of loci: 11q13 and 13q32; 22q13 and 21q22; 12q24 and 16p13; 8q24 and 7q21; 20p13 and 16q21; and 5p13 and 16p12.

Tomlins et al. (2007) explored the mechanism of ETV1 outlier expression in human prostate tumors and prostate cancer cell lines. They identified previously unknown 5-prime fusion partners in prostate tumors with ETV1 outlier expression, including untranslated regions from a prostate-specific androgen-induced gene (SLC45A3; 605097) and an endogenous retroviral element, HERV-K_22q11.23, a prostate-specific androgen-repressed gene (C15ORF21; 611314), and a strongly expressed housekeeping gene (HNRNPA2B1; 600124). To study aberrant activation of ETV1, Tomlins et al. (2007) identified 2 prostate cancer cell lines that had ETV1 outlier expression. Through distinct mechanisms, the entire ETV1 locus (7p21) is rearranged to a 1.5-Mb prostate-specific region at 14q13.3-q21.1 in both cell lines, in one by cryptic insertion and in the other by balanced translocation. Because the common factor of these rearrangements is aberrant ETV1 overexpression, Tomlins et al. (2007) recapitulated this event in vitro and in vivo, demonstrating that ETV1 overexpression in benign prostate cells and in mouse prostate confers neoplastic phenotypes. Identification of distinct classes of ETS gene rearrangements demonstrated that dormant oncogenes can be activated in prostate cancer by juxtaposition to tissue-specific or ubiquitously active genomic loci.

Using dual-color FISH in LNCaP prostate cancer cells, which are androgen-sensitive but lack the TMPRSS2-ERG fusion gene, Mani et al. (2009) observed that stimulation with the AR ligand dihydrotestosterone (DHT) for 60 minutes induced proximity between the TMPRSS2 and ERG genomic loci. The effect was dependent upon AR, as the same proximity was not induced in an androgen-insensitive prostate cancer cell line. To determine whether the induced proximity facilitates formation of these gene fusions, Mani et al. (2009) treated LNCaP cells with DHT for 12 hours and then irradiated the cells to induce DNA double-strand breaks. TMPRSS2-ERG fusions were detected in 25% of clones treated with 3-Gy irradiation but in only 2.3% of those treated with 1-Gy. Mani et al. (2009) speculated that androgen signaling colocalizes the 5- and 3-prime gene fusion partners, thereby increasing the probability of a gene fusion when subjected to agents that cause DNA double-strand breaks.

The TMPRS22 and ERG genes are arranged tandemly on chromosome 21q22. The TMPRSS2/ERG fusion joins TMPRSS2 exons 1 or 2 usually to ERG exons 2, 3 or 4, which results in activation of the ERG transcription factor. This fusion separates the ERG 3-prime centromeric regions from the 5-prime telomeric ends; deletions of this region can also occur. Attard et al. (2008) performed FISH studies of the TMPRS22/ERG genes in 445 prostate cancers from patients who had been managed conservatively. The authors identified an alteration, called 2+Edel, characterized by duplication of the TMPRS22/ERG fusion (detected as duplication of 3-prime ERG sequence) together with interstitial deletion of 5-prime ERG sequences. The alteration was found in 6.6% of cancers and was associated with very poor clinical outcome compared to cancers with normal ERG loci (25% vs 90% survival at 8 years). Cancers with 1 copy of 3-prime ERG (1Edel) did not have a worse clinical outcome. The findings were consistent with the hypothesis that overexpression of ERG that results from the fusion of 5-prime TMPRSS2 to 3-prime ERG is responsible for driving cancer progression. Attard et al. (2008) suggested that determination of ERG gene status, including duplication of the fusion of TMPRSS2 to ERG sequences in 2+Edel, may allow stratification of prostate cancer into distinct survival categories.

Berger et al. (2011) presented the complete sequence of 7 primary human prostate cancers and their paired normal counterparts. Several tumors contained complex chains of balanced (i.e., 'copy-neutral') rearrangements that occurred within or adjacent to known cancer genes. Rearrangement breakpoints were enriched near open chromatin, androgen receptor (AR; 313700), and ERG DNA binding sites in the setting of the ETS gene fusion TMPRSS2/ERG, but inversely correlated with these regions in tumors lacking ETS fusions. Berger et al. (2011) suggested that this observation suggests a link between chromatin or transcriptional regulation and the genesis of genomic aberrations. Three tumors contained rearrangements that disrupted CADM2 (609938), and 4 harbored events disrupting either PTEN (unbalanced events), a prostate tumor suppressor, or MAGI2 (606382) (balanced events), a PTEN-interacting protein not previously implicated in prostate tumorigenesis. Berger et al. (2011) concluded that genomic rearrangements may arise from transcriptional or chromatin aberrancies and engage prostate tumorigenic mechanisms.

Population Genetics

Seidman et al. (1985) estimated that 9% of white males and 10% of black males in the US will develop clinical prostate cancer in their lifetime. Silverberg (1987) stated that in American men, prostate cancer is the most common malignancy and the second most common cause of cancer deaths. Over 100,000 cases of prostate cancer had occurred annually in the United States in previous years, with 20,000 of these cases occurring in men under the age of 65 years.

In Los Angeles County, California, prostate cancer incidence differs considerably among the various racial-ethnic groups: highest in African Americans (116 per 1,000 person-years), intermediate in non-Hispanic whites (71 per 100,000 person-years), and lowest among Asians (Japanese at 39 per 100,000 person-years and Chinese at 28 per 100,000 person-years).

In 940 Ashkenazi Israelis with prostate cancer, Giusti et al. (2003) tested DNA obtained from paraffin sections for the 3 Jewish founder mutations: 185delAG (113705.0003) and 5382insC (113705.0018) in BRCA1 and 6174delT (600185.0009) in BRCA2. They estimated that there is a 2-fold increase in BRCA mutation-related prostate cancer among Ashkenazi Israelis. No differences were noted between the histopathologic features of cases with or without founder mutations, and no difference was found in the mean age at diagnosis between cases with or without a founder mutation.

Pathogenesis

Karhadkar et al. (2004) found that activity of the hedgehog (see 600725) signaling pathway, which has essential roles in developmental patterning, was required for regeneration of prostate epithelium, and that continuous pathway activation transformed prostate progenitor cells and rendered them tumorigenic. Elevated pathway activity furthermore distinguished metastatic from localized prostate cancer, and pathway manipulation modulated invasiveness and metastasis. Pathway activity was triggered in response to endogenous expression of hedgehog ligands, and was dependent upon the expression of Smoothened (601500), which is not expressed in benign prostate epithelial cells. Karhadkar et al. (2004) concluded that monitoring and manipulating hedgehog pathway activity may offer significant improvements in diagnosis and treatment of prostate cancers with metastatic potential.

Seligson et al. (2005) used immunohistochemical staining of primary prostatectomy tissue samples to determine the percentage of cells that stained for the histone acetylation and dimethylation of 5 residues in histones H3 (see 142780) and H4 (see 602822). Grouping of samples with similar patterns of modifications identified 2 disease subtypes with distinct risks of tumor recurrence in patients with low-grade prostate cancer. These histone modification patterns were predictors of outcome independently of tumor stage, preoperative prostate-specific antigen levels, and capsule invasion. Thus, Seligson et al. (2005) concluded that widespread changes in specific histone modifications indicate previously undescribed molecular heterogeneity in prostate cancer and might underlie the broad range of clinical behavior in cancer patients.

Luo et al. (2007) examined IKK-alpha (CHUK; 600664) involvement in prostate cancer and its progression. They demonstrated that a mutation that prevents IKK-alpha activation slowed down prostate cancer growth and inhibited metastatogenesis in TRAMP mice, which express SV40 T antigen in the prostate epithelium. Decreased metastasis correlated with elevated expression of the metastasis suppressor Maspin (154790), the ablation of which restored metastatic activity. IKK-alpha activation by RANK ligand (RANKL; 602642) inhibited Maspin expression in prostate epithelial cells, whereas repression of Maspin transcription required nuclear translocation of active IKK-alpha. The amount of active nuclear IKK-alpha in mouse and human prostate cancer correlated with metastatic progression, reduced Maspin expression, and infiltration of prostate tumors with RANKL-expressing inflammatory cells. Luo et al. (2007) proposed that tumor-infiltrating RANKL-expressing cells lead to nuclear IKK-alpha activation and inhibition of Maspin transcription, thereby promoting the metastatic phenotype.

Using a combination of high-throughput liquid and gas chromatography-based mass spectrometry, Sreekumar et al. (2009) profiled more than 1,126 metabolites across 262 clinical samples related to prostate cancer (42 tissues and 110 each of urine and plasma). These unbiased metabolomic profiles were able to distinguish benign prostate, clinically localized prostate cancer, and metastatic disease. Sarcosine, an N-methyl derivative of the amino acid glycine, was identified as a differential metabolite that is highly increased during prostate cancer progression to metastasis and can be detected noninvasively in urine. Sarcosine levels were also increased in invasive prostate cancer cell lines relative to benign prostate epithelial cells. Knockdown of glycine-N-methyl transferase (606628), the enzyme that generates sarcosine from glycine, attenuated prostate cancer invasion. Addition of exogenous sarcosine or knockdown of the enzyme that leads to sarcosine degradation, sarcosine dehydrogenase (604455), induced an invasive phenotype in benign prostate epithelial cells. Androgen receptor (AR; 313700) and the ERG (165080) gene fusion product (see 165080) coordinately regulate components of the sarcosine pathway. Sreekumar et al. (2009) concluded that by profiling the metabolomic alterations of prostate cancer progression, they revealed sarcosine as a potentially important metabolic intermediary of cancer cell invasion and aggressivity.

Ammirante et al. (2010) found that prostate cancer progression is associated with inflammatory infiltration and activation of IKK-alpha (600664), which stimulates metastasis by an NF-kappa-B (see 164011)-independent cell-autonomous mechanism (Luo et al., 2007). Ammirante et al. (2010) showed that androgen ablation causes infiltration of regressing androgen-dependent tumors with leukocytes, including B cells, in which IKK-beta (603258) activation results in production of cytokines that activate IKK-alpha and STAT3 (102582) and prostate cancer cells to enhance hormone-free survival.

Goldstein et al. (2010) showed that basal cells from primary benign human prostate tissue could initiate prostate cancer in immunodeficient mice. The cooperative effects of AKT (164730), ERG, and AR in basal cells recapitulated the histologic and molecular features of human prostate cancer, with loss of basal cells and expansion of luminal cells expressing PSA and alpha-methylacyl-CoA racemase (AMACR; 604489). Goldstein et al. (2010) concluded that the histologic characterization of cancers does not necessarily correlate with the cellular origins of the disease.

Studying mouse models, Ku et al. (2017) demonstrated that Rb1 (604041) loss facilitates lineage plasticity and metastasis of prostate adenocarcinoma initiated by Pten (601728) mutation. Additional loss of Tp53 caused resistance to antiandrogen therapy. Gene expression profiling indicated that mouse tumors resemble human prostate cancer neuroendocrine variants; both mouse and human tumors exhibited increased expression of epigenetic reprogramming factors such as Ezh2 and Sox2 (184429). Clinically relevant Ezh2 inhibitors restored Ar expression and sensitivity to antiandrogen therapy. Ku et al. (2017) concluded that their findings uncovered genetic mutations that enable prostate cancer progression, identified mouse models for studying prostate cancer lineage plasticity, and suggested an epigenetic approach for extending clinical responses to antiandrogen therapy.

Molecular Genetics

Using cDNA microarrays, Dhanasekaran et al. (2001) examined gene expression profiles of more than 50 normal and neoplastic prostate specimens and 3 common prostate cancer cell lines. Signature expression profiles of normal adjacent prostate, benign prostatic hypertrophy, localized prostate cancer, and metastatic, hormone-refractory prostate cancer were determined. Dhanasekaran et al. (2001) established many associations between genes and prostate cancer. They assessed 2 of these genes, hepsin (142440), a transmembrane serine protease, and PIM1 (164960), a serine/threonine kinase, at the protein level using tissue microarrays consisting of over 700 clinically stratified prostate cancer specimens. Expression of hepsin and PIM1 proteins was significantly correlated with measures of clinical outcome.

To explore potential molecular variation underlying the broad range of clinical behavior of prostate cancer from relatively indolent to aggressive metastatic disease, Lapointe et al. (2004) profiled gene expression in 62 primary prostate tumors, as well as 41 normal prostate specimens and 9 lymph node metastases, using cDNA microarrays containing approximately 26,000 genes. Unsupervised hierarchical clustering readily distinguished tumors from normal samples, and further identified 3 subclasses of prostate tumors based on distinct patterns of gene expression. High-grade and advanced stage tumors, as well as tumors associated with recurrence, were disproportionately represented among 2 of the 3 subtypes, 1 of which also included most lymph node metastases. To characterize further the clinical relevance of tumor subtypes, Lapointe et al. (2004) evaluated as surrogate markers 2 genes differentially expressed among tumor subgroups using immunohistochemistry on tissue microarrays representing an independent set of 225 prostate tumors. Positive staining for MUC1 (158340), a gene highly expressed in the 2 subgroups with 'aggressive' clinicopathologic features, was associated with an elevated risk of recurrence (P = 0.003), whereas strong staining for AZGP1 (194460), a gene highly expressed in the third subgroup, was associated with a decreased risk of recurrence (P = 0.0008). In multivariate analysis, MUC1 and AZGP1 staining were strong predictors of tumor recurrence independent of tumor grade, stage, and preoperative levels of prostate-specific antigen (PSA; 176820). The results suggested that prostate tumors can be classified according to their gene expression patterns.

Barbieri et al. (2012) sequenced the exomes of 112 prostate tumor and normal tissue pairs. New recurrent mutations were identified in multiple genes, including MED12 (300188) and FOXA1 (602294).

Pritchard et al. (2016) screened 20 DNA repair genes for germline mutation in 692 men with metastatic prostate cancer and identified 84 pathogenic mutations in 82 men (11.8%). This frequency was significantly higher than the rate of germline mutations in these genes in men with localized prostate cancer (4.6% among 499 men, p less than 0.001) or the prevalence among individuals in the ExAC browser (2.7% among 53,105 individuals without a known cancer diagnosis). Pathogenic variants were identified in 16 genes, including BRCA2 (600185) (37 men; 5.3%); ATM (607585) (11 men; 1.6%); CHEK2 (604373) (10 men; 1.9% of 534 men with data); BRCA1 (113705) (6 men, 0.9%), and RAD51D (602954) and PALB2 (610355) (each in 3 men, 0.4%).

Association with the SRD5A2 Gene on Chromosome 2p23

Nam et al. (2001) investigated the val89-to-leu polymorphism (V89L) of the SRD5A2 gene (607306) in 320 men undergoing evaluation for prostate cancer, and found that the adjusted odds ratio for having prostate cancer for men with at least 1 V allele was 2.53 compared to men with the L/L genotype (95% CI, 1.11-5.78; p = 0.03). When stratified for ethnic background, the effect of the V89L polymorphism did not vary among whites, blacks, and Asians. In a separate cohort of 318 men with known prostate cancer, Nam et al. (2001) found that the odds ratio for progression for men with at least 1 V allele was 3.32 compared to men with the L/L genotype (95% CI, 1.67-6.62; p = 0.0006).

In a genotype/haplotype association study involving a case-control sample of 1,117 brothers from 506 sibships with prostate cancer, Loukola et al. (2004) detected positive associations between prostate cancer risk and 2 single-nucleotide polymorphisms (SNPs) in SRD5A2 (V89L and -3001G-A) and observed an inverse association between high aggressiveness of prostate cancer and SRD5A2_Hap3. The authors could not confirm a previously reported association between prostate cancer and the SRD5A2 A49T polymorphism (607306.0012).

Association with the LMTK2 Gene on Chromosome 7q21-q22

For discussion of a possible association of prostate cancer with mutation in the LMTK2 gene (610989), see HPC4 (608658).

Association with the CYP3A4 Gene on Chromosome 7q22

Loukola et al. (2004) detected associations between prostate cancer risk or aggressiveness and a number of CYP3A4 (124010) SNPs and a CYP3A4 haplotype; they noted that both the CYP3A4*1B allele and the CYP3A4_Hap4 haplotype were inversely associated with low disease aggressiveness.

Association with the KLF6 Gene on Chromosome 10p14

Narla et al. (2001) identified loss of heterozygosity of 1 KLF6 (602053) allele in 77% of primary prostate tumors. Sequence analysis of the retained KLF6 allele revealed mutations in 71% of these tumors. See 602053.0001-602053.0005. Functional studies confirmed that whereas wildtype KLF6 upregulates p21 (WAF1/CIP1; 116899) in a p53-independent manner and significantly reduces cell proliferation, tumor-derived KLF6 mutants do not.

Association with the MXI1 Gene on Chromosome 10q25

Eagle et al. (1995) demonstrated that mutation in the MXI1 gene (600020) is involved in either the pathogenesis or the neoplastic evolution of some prostate cancer. The MXI1 protein negatively regulated MYC oncoprotein (190080) activity and thus potentially serves a tumor suppressor function. Furthermore, MXI1 maps to chromosome 10q25, a region that is deleted in some cases of prostate cancer. Eagle et al. (1995) detected mutations in the retained MXI1 allele in 4 primary prostate tumors with 10q24-q25 deletions, thus satisfying the Knudson hypothesis.

Association with the CD82 Gene on Chromosome 11p11

For discussion of the possible role of the KAI1 (CD82) gene in suppression of metastatic potential in prostate cancer, see 600623.

Association with the CDKN1B Gene on Chromosome 12p13

Chang et al. (2004) analyzed the CDKN1B gene (600778) in 188 families with hereditary prostate cancer and found a significant association between the SNP -79C/T (rs34330) and prostate cancer. The -79C allele was overtransmitted from parents to affected offspring, an association that was observed primarily in offspring whose age at diagnosis was less than 65 years. Chang et al. (2004) suggested that germline variants of this gene play a role in prostate cancer susceptibility.

Association with the CDH1 Gene on Chromosome 16q22

In a Swedish population, Jonsson et al. (2004) demonstrated an association between the -160C/A promoter polymorphism in the CDH1 gene (192090.0018) and risk of hereditary prostate cancer. In an independent replication study population, Lindstrom et al. (2005) confirmed the association.

Association with the HNF1B Gene on Chromosome 17q12

For discussion of a possible association of prostate cancer with variation in the HNF1B gene (189907) gene, see HPC11 (611955).

Association with the ZNF652 Gene on Chromosome 17q21

In a genomewide association study of 3,425 African Americans with prostate cancer and 3,290 African American controls with follow-up in 1,844 cases and 3,269 controls of African ancestry, Haiman et al. (2011) identified a risk variant on chromosome 17q21 (rs7210100, odds ratio per allele of 1.51, p = 3.4 x 10(-13) for the combined cohorts). The frequency of the risk allele was about 5% in men of African descent, whereas it was rare in other populations (less than 1%). The variant rs7210100 is located in intron 1 of the ZNF652 gene (613907).

Association with the SPOP Gene on Chromosome 17q21

Barbieri et al. (2012) sequenced the exomes of 112 prostate tumor and normal tissue pairs. SPOP (602650) was the most frequently mutated gene, with mutations involving the SPOP substrate-binding cleft in 6 to 15% of tumors across multiple independent cohorts. Prostate tumors with mutant SPOP lacked ETS family (see 164720) gene rearrangements and showed a distinct pattern of genomic alterations. Barbieri et al. (2012) concluded that SPOP mutations may define a novel molecular subtype of prostate cancer.

Zuhlke et al. (2014) identified a heterozygous missense mutation in the SPOP gene in a patient with familial prostate cancer showing linkage to chromosome 17 (602650.0001).

Association with the CHEK2 gene on Chromosome 22q12

For discussion of the role of the CHEK2 Gene in the development of prostate cancer, see 604373.

Association with the AR Gene on Chromosome Xq12

One of the critical functions of the product of the androgen receptor gene (AR; 313700) is to activate the expression of target genes. This transactivation activity resides in the N-terminal domain of the protein, which is encoded in exon 1 and contains polymorphic CAG and GGC repeats (microsatellites). A smaller size of the CAG repeat is associated with a higher level of receptor transactivation function, thereby possibly resulting in a higher risk of prostate cancer. Schoenberg et al. (1994) demonstrated contraction in this microsatellite from 24 to 18 CAG units in an adenocarcinoma of the prostate, and the affects of the shorter allele were implicated in the development of the tumor. Edwards et al. (1992) and Irvine et al. (1995) showed that the prevalence of short CAG alleles was highest in African American males with the highest risk for prostate cancer, intermediate in intermediate-risk non-Hispanic whites, and lowest in Asians at very low risk for prostate cancer. Irvine et al. (1995) found that high-risk African Americans also had the lowest frequency of the GGC allele. Consistent with the interracial variation in CAG and GGC distributions, there was an excess of white patients with short CAG repeats relative to white controls. Irvine et al. (1995) found a statistically significant negative association between the number of CAG and GGC repeats among the prostate cancer patients. Overall, the data were interpreted to suggest a possible association between the microsatellites of the AR gene and the development of prostate cancer.

Castration-Resistant Prostate Cancers

Grasso et al. (2012) sequenced the exomes of 50 lethal, heavily pretreated metastatic castration-resistant prostate cancers (CRPC) obtained at rapid autopsy (including 3 different foci from the same patient) and 11 treatment-naive, high-grade localized prostate cancers. Grasso et al. (2012) identified low overall mutation rates even in heavily treated CRPCs (2.00 per megabase) and confirmed the monoclonal origin of lethal CRPC. Integrating exome copy number analysis identified disruptions of CHD1 (602118) that define a subtype of ETS gene family fusion-negative prostate cancer. Similarly, Grasso et al. (2012) demonstrated that ETS2 (164740), which is deleted in one-third of CRPCs (commonly through TMPRSS2:ERG fusions), is also deregulated through mutation. Furthermore, they identified recurrent mutations in multiple chromatin- and histone-modifying genes, including MLL2 (602113) (mutated in 8.6% of prostate cancers), and demonstrated interaction of the MLL complex with the AR, which is required for AR-mediated signaling. Grasso et al. (2012) identified novel recurrent mutations in the AR collaborating factor FOXA1 (602294) in 5 of 147 (3.4%) prostate cancers (both untreated localized prostate cancer and CRPC), and showed that mutated FOXA1 represses androgen signaling and increases tumor growth. Proteins that physically interact with the AR, such as the ERG gene fusion product, FOXA1, MLL2, UTX (300128), and ASXL1 (612990), were found to be mutated in CRPC. Grasso et al. (2012) concluded that their study described the mutational landscape of a heavily treated metastatic cancer, identified novel mechanisms of AR signaling deregulated in prostate cancer, and prioritized candidates for future study.

Xu et al. (2012) found that the oncogenic function of EZH2 (601573) in cells of castration-resistant prostate cancer is independent of its role as a transcriptional repressor. Instead, it involves the ability of EZH2 to act as a coactivator for critical transcription factors including the androgen receptor. This functional switch is dependent on phosphorylation of EZH2 and requires an intact methyltransferase domain.

To elucidate mechanisms of castration resistance, Lunardi et al. (2013) performed an integrated analysis that leveraged data from treatment of genetic mouse models of prostate cancer with clinical data from patients. The authors found that castration counteracted tumor progression in a Pten loss-driven mouse model of prostate cancer through the induction of apoptosis and proliferation block. Conversely, this response was bypassed with deletion of either Trp53 (191170) or Zbtb7a (605878) together with Pten, leading to the development of castration-resistant prostate cancer. Mechanistically, the integrated acquisition of data from mouse models and patients identified the expression patterns of XAF1 (606717), XIAP (300079), and SRD5A1 (184753) as a predictive and actionable signature for castration-resistant prostate cancer. Lunardi et al. (2013) showed that combined inhibition of the XIAP, SRD5A1, and AR (313700) pathways overcomes castration resistance.

Using in vitro and in vivo human prostate cancer models, Mu et al. (2017) showed that prostate tumors can develop resistance to the antiandrogen drug enzalutamide by a phenotypic shift from androgen receptor-dependent luminal epithelial cells to androgen receptor-independent basal-like cells. This lineage plasticity is enabled