INTRODUCTION — Prostate cancer is the second most common cancer in men worldwide according to data from the GLOBOCAN database. If completely localized within the capsule of the gland, some favorable-risk prostate cancers can be managed with active surveillance or potentially cured by definitive local therapy (eg, radical prostatectomy, radiation therapy, brachytherapy). In contrast, non-organ-confined disease is often fatal. (See "Initial approach to low- and very low-risk clinically localized prostate cancer" and "Overview of systemic treatment for advanced, recurrent and metastatic castration-sensitive prostate cancer and local treatment for patients with metastatic disease".)
Here we will provide an overview of the molecular changes that are proposed to be involved in the initiation and progression of prostate cancer, particularly those that accompany metastatic disease and castration resistance. The diagnosis, staging, and treatment of prostate cancer are covered elsewhere. (See appropriate topic reviews.)
PATHOGENESIS — Metastatic prostate cancers can be lethal because they heterogeneously contain both androgen-dependent and androgen-independent malignant cells. For those cells that are androgen dependent, a critical level of androgen is required to activate a sufficient number of androgen receptors (ARs) so that transcription of death-signaling genes is repressed [1]. Androgen ablation therapies allow these genes to be expressed, triggering the biochemical cascade that results in apoptotic cell death, resulting in the eradication of the large fraction of androgen-dependent cancer cells [2,3]. In contrast, androgen ablation does not induce apoptosis in androgen-independent cells [4]; their eventual outgrowth is responsible for the lethality of advanced disease. (See "Overview of systemic treatment for advanced, recurrent and metastatic castration-sensitive prostate cancer and local treatment for patients with metastatic disease".)
Although prostate cancer typically presents in men over the age of 65, a growing body of evidence suggests that prostatic carcinogenesis is initiated much earlier [5]. Prostatic intraepithelial neoplasia (PIN) is thought to represent a precursor of adenocarcinoma, although not all cases progress to invasive disease. One of the most pressing clinical problems presented by prostate cancer is the difficulty in predicting its clinical course based on clinical or histologic features. (See "Precancerous lesions of the prostate: Pathology and clinical implications".)
Prostate cancer progression has been related to a number of genetic abnormalities that affect the AR and other molecules that are involved in the regulation of cell survival and apoptosis [6]. A multistep process of prostate carcinogenesis has been proposed in which progressive accumulation of genetic alterations is postulated to facilitate cellular transformation from normal prostate epithelium to PIN, invasive neoplasia, and castration resistance [6,7].
Over the last decade, a number of genes involved in or associated with prostate cancer have been identified and characterized [6]. Many of these genes or their protein products are under study for their value in clinical staging (sometimes termed "molecular staging") with the goal of more closely tailoring the selection of treatment to expected prognosis [8]. Perhaps more importantly, mechanistic studies to determine the biochemical result of specific molecular changes might reveal additional targets for therapy. Unfortunately, the heterogeneity of prostate cancers, the variability in analytic techniques, and the lack of suitable model systems for different stages of prostate progression have hampered progress. (See "Clinical presentation and diagnosis of prostate cancer".)
CHROMOSOMAL ABERRATIONS AND CANDIDATE GENES — Techniques such as fluorescence in situ hybridization (FISH), comparative genomic hybridization, and microsatellite analysis have demonstrated losses or gains in various chromosomes that are associated with prostate cancer progression [9-12]. In many instances, the target gene or genes have been identified. Further advances in positional cloning technology and the information provided by the human genome project are expected to facilitate the identification of candidate genes within these chromosomal regions.
There are three categories of chromosomal changes that may play a role in the development and progression of prostate cancer: those that are associated with genetic predisposition (eg, 1q deletions), somatic alterations that result in amplification of presumed oncogenes (eg, 7p and 8q), and somatic alterations that result in the loss of function of presumed tumor suppressor genes (eg, 8p, 10q, 12q, 13q, and 17p). Here we will provide an overview of selected molecular changes that occur during prostate cancer development and progression, with a particular emphasis on their prognostic and therapeutic value. Specific details of these chromosomal alterations are beyond the scope of this review and are discussed more in depth elsewhere [13,14].
Genetic factors that may influence prostate cancer risk are discussed elsewhere. (See "Risk factors for prostate cancer".)
Oncogenes — Oncogenes are homologs of normal cellular genes in which a mutational change results in constitutive activation and gain of function. These mutations either accelerate cellular growth and/or proliferation (eg, c-myc, beta catenin, human epidermal growth factor 2 [HER2], Ras), or facilitate abnormal cell division (eg, mitogen-activated protein kinase phosphatase 1 [MKP-1], B cell leukemia/lymphoma [bcl-2], telomerase).
c-myc — Gain of chromosome 8q sequences is a common feature of prostate cancer, particularly metastatic tumors [15]. The target gene is not known, but one candidate is c-myc, which is amplified in a subset of advanced tumors. The myc oncogene, located at chromosome 8q24, encodes a nuclear protein that is involved in the control of normal cell growth, differentiation, and apoptosis. Increasing levels of overexpression correlate with advancing stage and grade of prostate cancer [16-18], and high levels of amplification are found in a subset of men with androgen-independent disease [19-21].
MKP-1 — Several oncogenes involved in prostate carcinogenesis activate mitogen-activated protein (MAP) kinases, which can relay both proliferative (via extracellular regulated kinases [ERK]) and apoptotic signals (via Jun N-terminal protein kinases [JNK]) to the nucleus. MKP-1 is a dual-specificity MAP kinase phosphatase that dephosphorylates both phosphotyrosine and phosphothreonine residues on the JNK and ERK MAP kinases, providing overall protection from apoptotic influences. Overexpression of MKP-1 has been detected in the earliest stages of prostate cancer, including prostatic intraepithelial neoplasia (PIN) [22,23]. In one report of 51 prostate cancers, MKP-1 was overexpressed in PIN lesions, but its expression decreased with higher histologic grade and advanced disease stage [22].
Bcl-2 — Overexpression of the bcl-2 oncogene by prostate cancers confers resistance to apoptosis [24], and upregulation of bcl-2 (particularly in conjunction with p53 mutations, see below) is a frequent and important step in the progression to advanced or castration-resistant disease [25-29]. Bcl-2 tends to be expressed more strongly as malignant change progresses along the continuum from preinvasive to metastatic cancer [30-32]. As a result, the bcl-2 protein is a potential target for clinical intervention, particularly using antisense molecules such as oblimersen [33-35]. Its value as a prognostic marker is less certain [36-39].
Telomerase — Telomeres are specialized nucleoprotein structures that cap the ends of linear eukaryotic chromosomes [40]. Because of incomplete DNA replication at the ends of chromosomes, telomeres progressively shorten with each cell division in eukaryotic cells. This progressive loss of telomeres is postulated to be the "clock" by which normal cells senesce with age, and telomere maintenance appears to be an important regulator of cellular life span.
Telomerase, a ribonucleoprotein enzyme, compensates for telomere shortening during cell division by synthesizing telomeric DNA, thereby maintaining telomere length. In normal somatic cells, telomerase activity is usually undetectable, with the exception of hematopoietic cells, hair follicles, intestinal crypt cells, and epidermal basal cells. In contrast, upregulation of telomerase and amplification of telomeric DNA are detected in many cancers, including up to 90 percent of prostate cancers, and in high-grade PIN [41-47]. Higher-grade cancers appear to have maximally elevated telomerase activity [46,47].
Telomerase has been exploited both as a diagnostic tool and a therapeutic strategy in prostate cancer. A urinary assay for telomerase has been proposed as a noninvasive means of detecting prostate cancer, while the therapeutic potential of agents targeting telomerase and telomere-associated enzymes is beginning to be explored [48].
Beta catenin — The catenin family of proteins is thought to play a critical role in cell-cell adhesion. However, the role of beta catenin in prostate tumorigenesis appears to be distinct from cell adhesion. Although most cases of prostate cancer show downregulation of beta catenin at the messenger RNA (mRNA) and protein level [49,50], approximately 5 percent contain activating mutations within the beta catenin gene [51,52]. This gain of function mutation results in protein stabilization and accumulation in the nucleus, where beta catenin forms a transcriptional complex that is capable of upregulating target genes, such as the androgen receptor (AR) [51,53]. Up to 20 percent of lethal prostate cancers contain nuclear beta catenin, suggesting a more common role in advanced disease [54].
Ras pathway — The Ras oncogene is thought to be important in pathogenesis in many human malignancies (eg, non-small cell lung cancer).
In prostate cancer, Ras amplification does not seem to play a major role in disease initiation and development in American men [55,56], although somewhat higher mutation rates are reported in Japanese men [57,58].
In contrast, epigenetic inactivation of the RASSF1A gene (located on chromosome 3p21) via promoter methylation is present in approximately 60 to 70 percent of prostate cancers, particularly more-aggressive tumors [59-61].
An alternative mechanism for prostate cancer pathogenesis involves the DAB2IP gene, which encodes a protein that is a negative regulator of the Ras signal transduction pathway. A polymorphism of DAB2IP was associated with aggressive prostate cancer in an exploratory genome-wide association study [62].
HER2 — The protein product of the HER2 oncogene (also known as c-erbB-2) is a receptor tyrosine kinase belonging to the epidermal growth factor receptor (EGFR) family. Its role in prostate tumorigenesis is controversial. HER2 protein overexpression is detected in 20 to 25 percent of hormone-naive primary tumors and in 60 to 78 percent of cancers following androgen ablation [63,64]. However, the biologic significance of this finding is unclear. With the exception of one study [65], high-level amplification of the gene, representing the predominant mechanism of HER2 pathway activation (at least in breast cancer), is not observed in prostate cancer [19,63,64,66,67]. (See "HER2 and predicting response to therapy in breast cancer".)
EGFR pathway — EGFR, also known as HER1 or c-erbB-1, belongs to a family of receptors that includes the HER2 protein (c-erbB-2), as well as HER3 and HER4. EGFR is a transmembrane glycoprotein whose intracellular domain includes a tyrosine kinase, which is a therapeutic target in a number of other malignancies.
A possible role for EGFR in prostate cancer pathogenesis was suggested by a study of 2497 prostate cancers for whom follow-up data was available [68]. Overall, EGFR was expressed in 18 percent of cases. EGFR expression was associated with higher-grade tumors, more-advanced stage, and a decrease in recurrence-free survival.
Fatty acid synthase — A common occurrence in many cancers is the increased synthesis of long-chain fatty acids. Fatty acid synthase (FASN) is a key enzyme in this process. In prostate cancers, FASN expression is thought to function as an oncogene by preventing apoptosis. This is documented by the increased cell proliferation and tumorigenic growth of FASN-expressing prostate cell lines and transgenic models, as well as in human prostate specimens [69].
HOXB13 — The homeobox B13 (HOXB13) gene codes for a transcription factor that is important in prostate development. Sequencing studies have identified the G84E variant as associated with hereditary prostate cancer in some families [70]. The mechanism by which HOXB13 acts is not yet understood, but an understanding of this pathway may provide additional insights into sporadic cases of prostate cancer as well. (See "Genetic risk factors for prostate cancer".)
Oncogenes as a result of a gene fusion event — A recurrent gene fusion as a result of a chromosomal rearrangement has been identified in prostate cancer [71]. Fusion of the androgen-regulated gene TMPRSS2 and the ETS transcription family members (ERG, ETV1, ETV4, and ETV5) has been described as one of the most common recurrent genomic alterations in prostate cancer [72]. Such TMPRSS2-ERG fusion products are found in nearly one half of prostate-specific antigen (PSA)-screened prostate cancer patients and are absent in biopsies from benign lesions [73-75]. Furthermore, TMPRSS2-ERG fusions show a high correlation with disease recurrence after surgery for localized prostate cancer [76]. Since this initial observation, other permutations of gene fusions have been found, all of which involve an ETS transcription family member fused to a number of partners [72].
Tumor suppressor genes — In contrast to oncogenes, tumor suppressor genes exert their influence by controlling or inhibiting cell growth. When the gene mutates to cause a loss or reduction in function, this removes inhibitory influences on cell growth, thus allowing growth to proceed unchecked. As examples, the wild-type forms of genes such as GSTP1 protect cells from DNA damage, and the p53 gene product prevents cells from proliferating after damage has occurred; mutations in these putative tumor suppressor genes result in loss of these functions.
Deactivation of tumor suppressor genes can result from chromosomal deletion, loss of expression due to promoter hypermethylation, specific genetic mutations that result in deletion of a specific gene, or point mutations that result in the production of an aberrant protein product with reduced function.
PTEN/MMAC1 — The protein product of the phosphatase and tensin homologue (PTEN, also termed MMAC1) gene, located on chromosome 10q, negatively regulates the phosphoinositide 3-kinase (PI3-kinase)/Akt pathway. This pathway plays a critical role in a variety of growth factor signaling pathways that control the cell cycle and apoptosis. Although germline mutations of MMAC1 are found in numerous cancer predisposition syndromes [77], mutations in the PTEN gene have not been found in men with hereditary prostate cancer. (See "Overview of hereditary breast and ovarian cancer syndromes associated with genes other than BRCA1/2" and "Risk factors for prostate cancer".)
Nevertheless, PTEN clearly plays an important role in the progression of prostate cancer. PTEN alterations appear to be a late event, possibly influencing metastatic potential and progression to androgen independence rather than tumorigenesis [78,79]. As examples, high rates of loss of heterozygosity (LOH) for 10q are reported in advanced prostate cancer [80,81], and PTEN mutations are more commonly found in metastatic compared with localized prostate cancers (30 to 58 versus 5 to 27 percent of cases, respectively) [79,82-88]. A role for PTEN mutations in resistance to chemotherapy is also proposed [89], and gene therapy approaches using transfected wild-type PTEN are being developed [90].
Mxi1 — The product of the Mxi1 gene on chromosome 10q is a member of the helix-loop-helix-leucine zipper family and a potential negative regulator of c-myc (see 'Oncogenes' above). These proteins seem to be essential in cellular growth control and/or induction and maintenance of the differentiated state [91]. In prostate cancers, loss of Mxi1 expression, rather than point mutation, is the primary means of inactivation [80,92-94]. Mice lacking the Mix1 protein are susceptible to tumorigenesis, and Mxi1-deficient prostate epithelial cells have increased proliferative activity [91]. Others, using subtraction hybridization, have identified Mix-1 as one of eight target genes that are differentially expressed in androgen-dependent and androgen-independent prostate cancer xenografts [90].
GSTP1 — Hypermethylation of the p-class glutathione S-transferase (GSTP) gene promoter is possibly the most common genetic event in prostate cancer and appears to be an early event in tumorigenesis [95]. GSTPs are a family of enzymes that participate in detoxification by catalyzing the conjugation of many hydrophobic and electrophilic compounds with reduced glutathione; this protects the cells from environmental carcinogenic factors [96]. Over 90 percent of prostate carcinomas and 70 percent of high-grade PIN lesions have lost GSTP1 expression due to hypermethylation of the 5'-CpG island [97,98].
p53 — The p53 gene, located at chromosome 17p13.1, encodes a nuclear protein that acts as a transcription factor, blocking the progression of cells through the cell cycle late in the Gap 1 (G1) phase. The p53 gene plays a central role in the regulation of transcriptional events in the cell nucleus, particularly in the response to DNA-damaging agents, such as ionizing radiation and a variety of other carcinogens. The ability of wild-type p53 to prevent the proliferation of damaged cells has led to its description as the "guardian of the genome." A mutation in the p53 gene abrogates the G1 checkpoint and changes the cellular response to DNA damage, thus promoting genetic instability and the acquisition of additional mutations.
The significance of p53 mutations in prostate cancer is controversial. A number of studies using different analytical tools yield widely varying mutation rates among human prostate cancer specimens, a situation that is complicated by significant intratumoral heterogeneity within a single specimen [99]. Mutations in p53 have been reported in 60 to 90 percent of advanced prostate cancers [100], as well as in between 10 to 20 percent of primary tumors [101-105]. The reported incidence in PIN lesions ranges from 6 to 56 percent [105,106].
Despite these methodologic difficulties, accumulating data support a prognostic role for p53 mutations in human prostate cancer. The presence of p53 mutations in radical prostatectomy specimens is associated with increased proliferative rates, higher-grade disease, androgen independence, and a poor clinical outcome [17,25,29,36,39,101,102,107-115]. Furthermore, it has been proposed that p53 mutations in primary tumors are associated with increased metastatic potential [39,100,101,116]. In addition, abnormal p53 expression has been associated with higher rates of biochemical progression in men managed with radiation therapy and androgen deprivation therapy [117]. In a more recent study, there was a positive correlation between the expression of the proliferation marker Ki-67, p53, and the p53 regulator MDM2 and the development of distant metastases and overall mortality from prostate cancer [118].
In contrast, studies of p53 assessment in prostate needle biopsies have provided conflicting results, likely due to sampling variability, tumor heterogeneity, and multifocality [37,38,119,120]. A prospective multicenter study is needed to confirm the value of p53 assessment in needle biopsy prior to radical prostatectomy.
Although human trials have not begun, preclinical results suggest a potential therapeutic benefit for gene therapy using viral constructs that contain wild-type p53 in men with prostate cancer [121,122].
TGF-b1 — In normal prostate epithelial cells, transforming growth factor beta (TGF-b) induces apoptosis and inhibits proliferation. However, prostate tumors frequently do not express TGF-b receptors (type I and II), and as they progress, tumor cells shift to autocrine regulation of TGF-b. TGF-b overexpression coupled with the underexpression of TGF-b receptors allows prostate cancers to escape the growth inhibitory effect of TGF-b.
In prostate cancers, abnormalities in TGF-b are associated with higher histologic grade, a more malignant phenotype, and greater metastatic potential [123-127]. As an example, in one report of men undergoing prostatectomy for apparently localized disease, elevated preoperative plasma levels of TGF-b were a strong predictor of biochemical progression, presumably the result of occult metastatic disease at the time of surgery [128]. Preoperative levels of TGF-b have been incorporated into a nomogram to predict biochemical progression-free survival following radical prostatectomy, although measurement of preoperative levels is not routine in most centers [129].
Rb/p16 pathway — Chromosome 13q is one of the most frequently altered chromosomes in prostate cancer. In one study, 58 percent of primary tumors showed LOH for 13q, and allelic loss was even more frequent in metastatic lesions [130]. Two known tumor suppressor genes (retinoblastoma [Rb] and BRCA2) map to this region, although other putative tumor suppressor genes may also reside in this location [131].
The Rb gene, so named because deletions or mutations of this gene are critical in the pathogenesis of retinoblastoma, is also involved in the pathogenesis of a variety of other solid tumors. The Rb gene encodes for a nuclear protein (pRb1) that is an important regulator of the cell cycle during the G0/G1 phase; its activity is dependent on its phosphorylation status during the cell cycle. At the end of mitosis and during most of G0/G1, pRb1 is dephosphorylated, which actively suppresses the transition from G1 to synthesis (S) phase. Phosphorylation of pRb1 by cyclin-dependent kinases (CDKs) in late G1 phase leads to its inactivation, thereby leading to cell cycle progression. Multiple CDK inhibitors have been identified (eg, p16INK4A [also called CDKN2A or MTS-1, on chromosome 9p] and p21WAF1/CIP1) that block entry into S phase by decreasing pRb1 phosphorylation. This entire pathway of cell cycle control is often referred to as the Rb/p16 pathway.
In many solid tumors, dysregulation of the Rb/p16 pathway is an early event in tumorigenesis. However, the role of this pathway in prostate cancer is unclear. In early prostate cancer, deletion of p16 and Rb genes is reported in some, but not all, series, and the role of these genetic alterations in tumorigenesis and/or tumor progression remains undefined [132-140]. Other studies suggest that mutations in the Rb gene, and mutation or promoter methylation of the CDKN2A gene are more common in advanced prostate cancer [134,141-147]. Preclinical gene therapy studies using adenoviral vectors containing wild-type p16 or Rb genes are underway [148,149].
Although p21 is a growth inhibitory molecule, most series report expression in 25 to 30 percent of primary prostate tumors [150-153]. Furthermore, overexpression appears to be associated with progression toward androgen-independent disease [150]. In vitro studies of antisense strategies directed against p21 show early promise [154].
CDKI1B — The CDK inhibitor 1B gene (CDKI1B, previously called p27/Kip1) maps to 12p12-13.1, a region that is frequently deleted in advanced prostate cancer [155]. CDKI1B is an inhibitor of CDK4; as such, it is a negative regulator of cell cycle initiation. In regards to prostate cancer, several clinical studies have indicated a loss of CDKI1B protein expression in prostate cancers and PIN, indicating this as a potentially early event [156-159]. Furthermore, loss of CDKI1B protein expression in radical prostatectomy specimens has been shown to be an adverse prognostic factor in patients with clinically localized prostate cancer [143,157,160-163] and in those undergoing salvage prostatectomy after radiation therapy failure [164]. Because of this, loss of CDKI1B has been suggested as a component of molecular staging [162,165]. As with other putative tumor suppressor genes, preclinical studies with recombinant adenovirus expressing CDKI1B are underway [166].
Nkx3.1 — Nkx3.1 is an androgen-regulated prostate-specific homeobox gene on chromosome 8q that appears to be critical in the development of the normal prostate [167,168]. In some [169-171], but not all studies [168,172], decreased expression of Nkx3.1 is associated with both PIN and invasive prostate cancer. Others have shown a correlation between abnormalities in Nkx3.1 expression, and advanced-grade and castration-resistant disease [171-173].
KLF6 — Kruppel-like factor 6 (KLF6) is a zinc finger transcription factor. Its possible role as a tumor suppressor gene in prostate cancer was suggested in one series in which LOH for this allele was shown in 17 of 22 primary prostate tumors [174]. Further study of this gene and its role in prostate cancer is needed.
Metastasis suppressor genes — Advanced prostate carcinoma is characterized by local invasion, spread to regional lymph nodes, and more distant osseous metastases. Much effort has been directed toward the identification and characterization of genes that are involved in metastatic behavior, both for the purpose of obtaining prognostic indicators for aggressive disease, and to understand why and how prostate cancer preferentially metastasizes to bone. (See "Bone metastases in advanced prostate cancer: Clinical manifestations and diagnosis", section on 'Pathophysiology'.)
The identification of several genes that appear to function as metastasis suppressor genes has provided important insight into factors regulating metastatic spread and growth. By definition, metastasis suppressor genes specifically prevent the formation of overt metastasis without affecting the formation or growth of primary tumors [175,176].
The technique of microcell-mediated chromosomal transfer (MMCT) has been the most commonly used method to identify putative metastasis suppressor genes in prostate cancer. It is based on the work of Ichikawa et al, who demonstrated that fusion of nonmetastatic and highly metastatic Dunning rat prostate carcinoma cells produced nonmetastatic hybrid cells that were unaffected in terms of tumorigenicity and in vitro growth rates [177]. Subsequent studies suggested that the loss of specific chromosomes was responsible for the metastatic phenotype. MMCT was then used to selectively introduce single, tagged human chromosomes into metastatic recipient cells, thereby permitting the localization of metastasis suppressor genes to specific human chromosomes. Two of the genes identified in the rat Dunning model (KAI1 and CD44) appear to suppress metastatic potential at an early stage of the metastatic cascade.
KAI-1/CD82 — KAI1 was the first prostate cancer metastasis suppressor gene identified using MMCT; it maps to chromosome 11p11.2, and downregulation of this gene has been found in advanced prostate tumors and metastases [178,179]. The KAI1 gene encodes a membrane glycoprotein that is widely expressed, but its precise biochemical function in metastasis suppression has not been clearly defined. In vitro studies suggest that it may play a role in integrin signaling, cell adhesion, and motility [180-182].
E-cadherin (16q22) — E-cadherin (CDH1) is a Ca2+ dependent homotypic cell adhesion molecule that is implicated as a tumor/invasion suppressor in human carcinomas; it also functions as a binding partner for beta catenin, which is overexpressed in advanced prostate cancers. (See 'Oncogenes' above.)
Aberrant E-cadherin expression appears to be associated with high histologic grade and more aggressive tumor behavior [183-186], and deletions of chromosome 16q and/or loss of E-cadherin function are common in prostate cancer metastases [187,188]. Furthermore, using MMCT, 16q has been strongly associated with suppression of metastatic ability [189].
Additional evidence supporting the role of E-cadherin as a prognostic marker comes from the observation that decreased expression of E-cadherin and increased expression of N-cadherin are associated with adverse clinicopathologic features [190].
ACQUIRED CASTRATION RESISTANCE — Although the majority of men have a favorable initial response to androgen ablation, the development of hormone resistance is inevitable and can be responsible for the lethality of advanced prostate cancer. Multiple mechanisms have been postulated to account for the acquisition of castration resistance. A detailed review of the specific molecular changes that are associated with castration resistance is beyond the scope of this discussion, although these changes are well reviewed elsewhere [14,191]. Here we will provide a brief overview of some of the mechanisms that are thought to underlie the development of castration resistance.
In view of the importance of androgens to the growth of prostate cancer, it is reasonable to ask if alterations in the androgen receptor (AR) could be responsible for androgen-independent growth [192]. The AR gene is overexpressed in approximately 20 to 30 percent of hormone-refractory prostate cancers [193,194], although more sensitive techniques, such as gene expression profiling, suggest that amplification is nearly universal when hormone-resistant human prostate cancer cell lines are compared with the hormone-sensitive tumors from which they were derived [195]. In theory, such alterations might possibly enhance the response of the cancer cell to low levels of endogenous androgens [196,197].
Activating mutations in the AR gene have also been reported [198-200]. Such mutations in the steroid binding domain may alter ligand specificity, allowing other nonandrogenic steroids, as well as antiandrogens, to bind and activate the mutant AR, even when the level of systemic androgen is fully suppressed. AR mutations have been most intensely studied in the context of the antiandrogen withdrawal response, in which acquired mutations in the AR are thought to result in a paradoxical stimulatory effect of antiandrogens on prostate tumor cells. Mutations in the hormone binding domain of the AR have been detected in a small fraction of bone marrow specimens of men with castration-resistant prostate cancer [199-201]. In functional studies, these mutant receptors were stimulated rather than inhibited by some, but not all, antiandrogens [202]. However, other data suggest that AR mutations are not responsible for the antiandrogen withdrawal response, and this remains a controversial area [200]. (See "Alternative endocrine therapies for castration-resistant prostate cancer", section on 'Antiandrogen withdrawal'.)
Other specific AR mutations and splice variants deleting the ligand binding domain provide an additional pathway for androgen escape in prostate cancer cells [203,204]. The predominant V7 AR splice variant has been strongly associated with castration resistance and overall poor patient survival [203]. (See "Castration-resistant prostate cancer: Treatments targeting the androgen pathway", section on 'Resistance'.)
Another mechanism for castration resistance growth involves molecular changes in several regulatory genes that may prevent the transcription or expression of death signaling genes after androgen ablation [205,206] or prevent the cancer cells from responding to these apoptotic influences through constitutive expression of genes that are located downstream in the apoptotic cascade (eg, upregulation of B cell leukemia/lymphoma [bcl-2], see above). Alternatively, resistance to androgen ablation might also involve "crosstalk" between the AR and other signaling pathways that are induced by peptide growth factors. In theory, androgen-independent AR activation could then be achieved via other pathways that are activated by specific peptide growth factor receptors [207,208].
Finally, molecular changes within the prostate cancer cells may result in the ability to synthesize and secrete specific ligands and their receptors, essentially creating an autocrine survival pathway that is independent of androgen signaling pathways [209]. One such example is the secretion of neurotrophin ligands and their trk receptors by hormone-resistant prostate cancer cells [210]. Small molecule inhibitors of these tyrosine kinase receptors can induce apoptosis in hormone-resistant prostate cancer cells [211,212], and these trk inhibitors are currently in clinical trials.
SUMMARY — Prostate cancer continues to be a major health problem despite significant advances in prevention and early detection. The mechanistic characterization of genes shown to play a functional role in the development and progression of prostate cancer is beginning to provide important insights into the nature and behavior of the disease. The goal of these studies is to identify additional targets for staging and treatment.
ACKNOWLEDGMENT — We are saddened by the death of Nicholas Vogelzang, MD, who passed away in September 2022. UpToDate gratefully acknowledges Dr. Vogelzang's role as Section Editor on this topic, and his dedicated and longstanding involvement with the UpToDate program.