(PIA), that might be a precursor to prostatic intraepithelial neoplasia (PIN) and to prostate cancer (5). Specifically, PIA lesions contain regions of highly proliferative, but blunted and dysfunctional prostatic secretory cells, that are present in a backdrop of chronic inflammatory leukocytes (5). PIA lesions are often found adjacent to and directly merging into regions of PIN and prostate cancer (6), giving rise to the notion that PIA lesions may be direct precursors to PIN and prostate cancer. The epithelial cells in these PIA lesions appear to be under tremendous stress as they express high levels of stress response proteins such as COX2, GSTA1, and GSTP1 (5, 7, 8). Indeed, this setting of inflammation, proliferation, and stress found in PIA lesions may provide the initial selection pressures for prostate epithelial cells to accumulate the somatic genome alterations necessary for carcinogenesis.

PIN lesions, which are often contiguous with regions of PIA, are characterized by the presence of malignant-appearing, proliferating prostate epithelial cells contained within a normal glandular architecture. PIN lesions have been considered to be immediate precursors of adenocarcinoma of the prostate because they are more frequently observed in prostates that also contain prostate cancers, and they often directly merge into regions of prostate adenocarcinoma (9). The atypical, malignant-appearing cells in these lesions may have already begun accumulating the somatic genome alterations characteristic of prostate cancer.

Like precursor PIN lesions, primary prostate cancers are most frequently observed in the peripheral zone of the prostate (9). Patients with adenocarcinoma of the prostate can harbor multiple, often heterogeneous, cancerous lesions. The aggressiveness, extent, and long-term treatment outcomes of prostate cancers can be estimated by nomograms that take into account the histological grade of the lesions, the clinical stage, and the serum PSA (10-12). Histological grading of prostate adenocarcinomas is specified by the Gleason score. The Gleason grade, or pattern, describes the differentiation and architectural patterns of prostate cancer, and is scaled from one, referring to well-differentiated lesions, to five, referring to poorly-differentiated lesions. In order to account for the often heterogeneous nature of prostate adenocarcinomas, the combined Gleason score is a sum of the two most prevalent Gleason patterns observed in any given prostate cancer lesion. Staging of prostate cancer follows the revised American Joint Commission on Cancer TNM conventions (13-15). One nomogram suggests that men undergoing radical prostatectomy for treatment of primary prostate cancer can be stratified into three groups: a low risk group (Gleason score < 6, AND stage of T1c to T2a, AND serum PSA < 10 ng/ml) with 83% 10-year recurrence free survival, an intermediate risk group (Gleason score = 7, or stage of T2b, or serum PSA between 10 and 20 ng/ml) with a 46% 10-year recurrence free survival, and a high risk group (Gleason score > 8, or stage T2c or higher, or serum PSA > 20 ng/ml) with a 26% 10-year recurrence free survival(12). Unfortunately, the use of these nomograms is limited by the variability in the combined Gleason score assigned by different pathologists evaluating the same prostate cancer lesion (16, 17). This remains a significant challenge despite the development of educational texts and tools on the internet by expert pathologists (18, 19). Another potential approach to prognostication would be the identification of molecular markers for prostate cancer risk stratification, either to directly predict post-treatment outcomes, or to predict prostate cancer stage and grade and therefore indirectly predict for post-treatment outcomes. Additionally, as we shall discuss in this chapter, identification of such markers may also provide insight into the molecular pathogenesis of prostate cancer progression and metastasis development.

Since survival and mortality from prostate cancer is directly related to the development of progressive metastatic disease, the prognostic markers for risk stratification of prostate cancer are often indirect measures of metastatic potential. Of note, since Gleason scores stratify prostate cancers based on architecture, they cannot be applied to characterize metastatic prostate cancer. One direct measure for the development of metastases is post-prostatectomy rise of serum PSA, termed PSA recurrence. In fact, PSA recurrence is so predictive for prostate cancer recurrence and development of metastases, that it is commonly used for follow-up after radical prostatectomy for treatment of primary prostate cancer. When metastases do develop, they can occur locally at pelvic lymph nodes, or involve distant organs, typically the axial or appendicular skeleton, and less commonly, the liver, lungs, and brain. Metastases to pelvic lymph nodes can lead to compression of iliac veins and edema of lower extremeties. Approximately 90% of advanced stage prostate cancer patients develop skeletal metastases, most typically at the lumbar spines or pelvic bones, leading to significant morbidity including severe bone pain, spinal cord compression, and pancytopenia due to invasion of the bone marrow (20, 21). Liver metastases can produce abdominal pain and jaundice in some rare cases. Lung metastases can lead to chest pain, coughing, as well as paraneoplastic syndromes due to ectopic hormone production from small cell forms of metastatic prostate cancers. Other uncommon manifestations of metastatic prostate cancer include malignant retroperitoneal fibrosis due to metastasis into the periureteral lymphatics, and disseminated intravascular coagulation (DIC) (9). The mechanisms guiding prostate cancer cells to metastasize so commonly to a specific set of tissues, such as bone, are largely unknown and the subject of intense research. In this chapter, we will examine how DNA methylation changes in metastatic prostate cancer lesions may help us understand how these metastases develop.

There is mounting evidence in the literature to view prostate cancer progression as a continuum from normal prostate, to PIA, to PIN, to primary prostate cancer, and finally to metastatic and androgen-independent metastatic prostate cancer (Figure 1). Likewise, within primary prostate cancers, there is a continuum of aggressiveness as measured by the combined Gleason score and tumor stage. Several recent studies have tracked the molecular changes at each step along this progression. A host of somatic genome changes and biochemical alterations have been implicated at each step along the progression, and have provided several clues to the pathogenesis of symptomatic, life-threatening, metastatic prostate cancer (Figure 1). These changes are often quite heterogeneous between different patients with prostate cancer, different cancer lesions within the same patient, and even different regions within the same cancer (22). One of the earliest and most frequent genome alterations in prostate cancer is the shortening of telomere repeat sequences at the ends of chromosomes in prostate cancers. Telomeres, which are repetitive sequences at the ends of chromosomes that protect against inappropriate loss and recombination of chromosomes during replication, are significantly shortened in PIN and prostate cancer lesions (23, 24). These shortened telomeres may allow illegitimate chromosomal recombination and genetic instability early in prostate carcinogenesis, leading to prostate cancer progression. Chromosomal gains at 7p, 7q, 8q, and Xq, and losses at 8p, 10q, 13q, and 16q are among the most commonly reported chromosomal abnormalities in prostate cancer. Somatic genome alterations and expression changes at specific genes within these chromosomal regions have been implicated in prostate cancer progression (22). For instance, NKX3.1, a prostate specific homeobox gene required for normal prostate development, is located on 8p21 (25-27). Loss of 8p21 and absence of NKX3.1 expression appear to be frequent changes early during prostate cancer progression, occurring as early as the precursor PIN lesions (28, 29). PTEN, a tumor suppressor gene located at chromosome 10q that encodes a phosphatase inhibitor of the phosphatidylinositol 3'-kinase/protein kinase B (PI3K/Akt) signaling pathway that is needed for cell cycle progression and cell survival, frequently contains somatic alterations in prostate cancer (30-34). Though it is expressed in normal prostates and PIN lesions, PTEN is under-expressed and contains somatic alterations in primary prostate cancer, and even more so in metastatic prostate cancer lesions (35-37). Additionally, somatic gene alterations at the androgen receptor (AR), such as gene amplifications and ligand-specificity altering mutations, have been documented in prostate cancer cells (38-41). These alterations may account for the inevitable development of androgen-insensitive cancers when patients with prostate cancer metastases are treated with androgen deprivation and/or anti-

androgen therapy (42-45). Metastatic prostate cancers can also acquire androgen-insensitivity by biochemical modifications, such as post-translational phosphorylation, on wild type AR causing constitutive activation even in the absence of androgen (46, 47). Development of hormone refractory metastatic prostate cancer carries an ominous prognosis as these patients have a median survival of only 9 - 12 months (48). Other biochemical alterations implicated in prostate cancers include gene expression changes. For instance, gene expression microarray experiments consistently document the over-expression of Hepsin and AMACR in prostate cancers, as well as EZH2 in metastatic androgen-independent prostate cancers (9). The precise role of these genes in prostate cancer progression has not yet been determined.

While the changes detailed above are strongly associated with prostate cancer, hypermethylation of CGI sequences is perhaps the earliest and most frequent somatic genome alteration in prostate carcinogenesis and progression. The remainder of this chapter will detail the specific DNA methylation changes that occur during each step of prostate cancer progression and use this information to build a potential model for metastasis development.

Figure 1. Summary of somatic genome changes occurring during prostate cancer progression.
10 Ways To Fight Off Cancer

10 Ways To Fight Off Cancer

Learning About 10 Ways Fight Off Cancer Can Have Amazing Benefits For Your Life The Best Tips On How To Keep This Killer At Bay Discovering that you or a loved one has cancer can be utterly terrifying. All the same, once you comprehend the causes of cancer and learn how to reverse those causes, you or your loved one may have more than a fighting chance of beating out cancer.

Get My Free Ebook

Post a comment