Homeobox D3 (HOXD3), is a gene whose silencing by methylation was shown to correlate with tumor grade [6]. By methylation-specific PCR, it was associated with biochemical recurrence in univariate analysis (P-value = 0.043) and showed evidence for interaction with pathological stage as a predictor variable in Cox regression analysis (P-value = 0.028) [6]. It is not being used as a commercial test currently.

Another new, related test is the Prostate Health index or PHI [11] from Beckman Coulter. The PHI is a new formula that combines three kallikreins (total PSA, free PSA, and p2PSA—the benign form of PSA or BPSA) into a single score that can be used to aid in clinical decision-making. Rhodes et al. [12] note that although the PHI test offers statistically significant results when applied to large populations, the clinical utility of the PHI test for individual patient management is not yet determined. The authors’ carefully worded conclusion states that “The PHI measurement, … may be useful to decrease unnecessary biopsies with improved specificity at various sensitivities for [prostate cancer] detection in men 50 years or older with PSA 2.0–10.0 ng/ml and negative DRE findings”.

PSA and its derivatives are kallikreins. [−2] proPSA (p2PSA) is a subtype of free PSA (the benign form of PSA), also called pro-PSA, and is estimated to be 2.5 times more specific in detecting prostate cancer in patients than a PSA screening [13]. It was approved by the US Food and Drug Administration (FDA) in June 2012 and has been available since late 2012.

Sarcosine is markedly elevated in metastatic prostate cancer and moderately elevated in the urine from primary prostate cancer and its biosynthetic enzyme is elevated in prostate cancer tissues [14]. Also, sarcosine was found to introduce invasion and intravasation of prostate cancer cells in an in vivo prostate cancer model. Prostarix, is a non-invasive urine test introduced in 2013 [15] that involves metabolic assessment using a quantitative liquid chromatography-mass spectroscopy method to accurately measure the urinary concentrations of four amino acids. This test is considered an alternative method of deciding whether to perform a set of prostate biopsies in men with mildly elevated serum PSA and negative digital rectal examination—in other words, men at equivocal risk.

In needle biopsy specimens, the prostate cancers for which a prognostic test is indicated are those with a highest Gleason score of 3 + 3 = 6 or 3 + 4 = 7. The test may tip the therapeutic response toward watchful waiting or cryotherapy rather than definitive therapy. However, cancer with a score of 4 + 3 = 7 or above is presumed aggressive and a test is not indicated. A number of American laboratories perform testing for the two most adverse prognostic markers in prostate pathology: PTEN loss and ERG expression. ERG is an androgen-responsive gene. Many studies have reproducibly shown that about 50 % of prostate cancer acquires reactivity for ERG, reflecting TMPRSS2-ERG rearrangement. About 16 % of high-grade prostatic intraepithelial neoplasia (HGPIN), the noninvasive precursor to invasive cancer [16] Carver] also is ERG-positive, and benign prostate never is. Tomlins et al. [17] showed that the most common rearrangement is TMPRSS2-ERG (21q22.2) followed by TMPRSS2-ETV1 (7p21.2), while TMPRSS2 (21q22)-ETV4 (17q21) (17q21) is rare (2 % of cases). Klezovich et al. [18] established a causal role of ERG overexpression in carcinogenesis. Overexpression of ERG in prostate cell lines increased cell invasion. Targeted expression in vivo in luminal prostate epithelial cells of transgenic mice resulted in development of focal precancerous prostatic intraepithelial neoplasia (PIN) [18] Klezovich]. Multifocal prostate cancer can show heterogeneity of ERG expression [19 Park].

ERG expression alone has only a marginal [20 Yoshimoto et al.] or no [21 Hoogland] effect on outcome as measured by biochemical recurrence-free survival. PTEN (Phosphatase and tensin homolog deleted on chromosome 10) is a key tumor suppressor gene in prostate cancer. Loss of expression of PTEN occurs in about one-third of prostate cancer [22 McCall].

PTEN deletions were present in 20.2 % (458/2,266) of prostate cancer (8.1 % heterozygous and 12.1 % homozygous [23 Krohn et al.]. PTEN deletions were linked to advanced stage, high Gleason grade, presence of lymph node metastasis, hormone-refractory disease, and presence of ERG gene fusion (all P < 0.0001). PTEN deletions were associated with biochemical recurrence in both univariate (P < 0.0001) and multivariate (P = 0.0158) analyses [24]. Given these findings, many private practice urologists, are electing to have PTEN loss study performed reflexively for their newly diagnosed cancers with Gleason scores of 3 + 3 = 6 or 3 + 4 = 7 in order to stratify the otherwise heterogeneous outcomes of these grade categories.

Lotan et al. [25] have investigated PTEN protein loss by immunohistochemistry to compare its assessment with that of FISH. PTEN immunostaining (IHC) detected up to 86 % of cases with PTEN genomic loss by comparison with single nucleotide polymorphisms. IHC was 87 % sensitive for hemizygous PTEN deletion, and 86 % sensitive for homozygous deletion. It was also found by IHC at times in the absence of apparent genomic loss. That is, in 37 % of cases in which SNP did not demonstrate PTEN loss and 45 % of cases in which FISH did not demonstrate PTEN loss, it was found by immunostaining. Others have reported similar findings in comparing IHC with FISH [26, 27]. Taken together, these data suggested that PTEN inactivation can occur by means other than genomic deletion, including insertions and epigenetic changes; and PTEN detection by IHC may hold some advantage over FISH. The main drawback, however, is that IHC cannot discriminate hemizygous and homozygous loss.

While ERG expression alone is not important alone, it becomes a significant prognostic factor only when combined with PTEN loss. Kaplan-Meier analysis shows that the status of these two markers can stratify men with prostate cancer into four significantly different groups [20]. Carver et al. [16] showed that in combination, ERG and PTEN define three groups of patients by prognosis: a good prognosis: ERG-/PTEN+ (29 % of cases); intermediate prognosis: ERG+/PTEN+ or ERG-/PTEN-(43 %); and poor prognosis: ERG+/PTEN- (28 %).

A special type of prostate cancer that is receiving abundant attention today is intraductal carcinoma (IDC) and the molecular biology peculiar to IDC has been reviewed [28]. IDC is an inherently high-grade type of prostate cancer, wherein duct spaces are distended to at least twice the diameter of a normal duct and contain a lumen-spanning proliferation of cells with marked nuclear enlargement, sometimes with luminal necrosis. The rates of ERG fusion and PTEN loss in IDC are at least commensurate with those of invasive carcinoma, and exceed the corresponding rates for HGPIN. Fine et al. [29] found no difference in ERG translocation/deletion or copy number increase in tumors from men with or without IDC. Immunoreactivity for ERG protein, reflecting a TMPRSS2-ERG gene fusion, has been reported in 35–75 % [30] of IDC but rarely or never in atypical cribriform lesions, which in the case of a topographic study using proximity to invasive cancer to discriminate IDC, [31] provides a molecular justification for topography being discriminatory. Schneider and Osunkoya found that the presence or absence of ERG reactivity in IDC always matched that of the acinar carcinoma; but the 35 % rate of expression of the invasive carcinoma associated with IDC was less than that not associated with IDC, suggesting that when IDC is present, the accompanying invasive component tends to have a unique phenotype [32] Schneider]. ERG immunostaining may be diagnostically useful in select cases. Cytoplasmic PTEN loss has been suggested as a marker to distinguish IDC from HGPIN, being observed in 84 % of IDC and 100 % of lesions intermediate between IDC and HGPIN, but never in HGPIN [28]. Nuclear reactivity for PTEN may be retained in IDC. IDC’s rate of PTEN loss exceeds the 35–45 % rates reported for acinar carcinoma [25, 33–36] but is similar to the significantly higher PTEN loss in Gleason grades 4–5 cancer [34], supporting IDC’s aggressiveness.

Is ERG immunostaining useful for diagnosis as well as prognosis? Andrews and Humphrey [2] showed that ERG reactivity is contributory only in uncommon cases of minimal adenocarcinoma when AMACR staining is negative or weak. Even then, in these cases ERG is informative in only a minority (29 %) of cases. Warrick and Humphrey [37] have also studied ERG expression in foamy gland carcinoma, which represents 17 % of cases. Sensitivity of AMACR for foamy gland carcinoma was 92 %, and sensitivity of ERG was 42 %. No AMACR-negative case was ERG positive. Low AMACR expression was detected in 24 of 72 cases (33 %), and of these cases, ERG was positive in 5 (21 %). This suggests that ERG is not useful as a first-round immunostain for foamy gland carcinoma and may only be useful as a supplementary marker in rare cases if AMACR is weak.

PSMA (FOLH1), has been studied as an immunohistochemical marker and has been found to predict progression in a statistically independent manner when controlling for tumor grade, stage, and clinical factors. In one study [38], PSMA expression and metastases to pelvic lymph nodes were significantly associated with time to PSA recurrence (HR, 1.4; 95 % confidence interval, 1.1–2.8, P = 0.017; and hazard ratio, 5; 95 % confidence interval, 2.6–9.7, P < 0.001, respectively). However, this marker is used mainly in a research capacity.

Survival pathways such as nuclear factor kappa B, MYC (c-MYC), phosphorylated AKT, its downstream effector FKHR (FOXO1) have all been studied in prostate cancer. Interestingly, all these markers correlated with the diameter of perineural invasion, in which a symbiotic relationship with nerves causes growth of both the cancer cells and the nerves [39 Olar et al]. These findings validate the importance of reporting perineural invasion when seen in prostatic biopsy cores.

Cell adhesion proteins mediate interactions of epithelial cells to each other and to the stroma, and thus cancer progression and metastasis. IDC shares in common with cribriform/papillary, large acinar (LA) cancer a dense concentration of neoplastic cells with absent intervening stroma and a uniquely adverse outcome. Invasive cribriform carcinoma, compared to other Gleason grade four patterns, was present in 61 % of men who experienced PSA failure but only 16 % of controls matched for follow-up duration and other clinicopathologic variables [40]. The quality of the stromal component of prostate cancer has been recognized as a prognostic factor. Indeed, loss of expression of integrins β1 in the epithelium and β3 in the stroma were noted in intraductal and cribriform cancer xenografts and untreated human tumors, compared with their non-cribriform counterparts [41]. This provides a rationale to expect that cribriform lesions—with or without surrounding basal cells—may be expected to show altered expression of other cell adhesion proteins.

Genomic Health, Inc. produces Oncotype Dx [42], a proprietary test of prostate cancer tissue for a series of genomic alterations pertaining to four pathways that drive prostate cancer. This furnishes a Genomic Prostate Score (GPS) that is said to predict outcome after surgery, regardless of what postoperative therapy is chosen. The original development studies looked at a subset of 185 prostate cancer patients who experienced biochemical recurrence after surgery.

Myriad Genetics markets a prognostic test for low (3 + 3 = 6) to intermediate (3 + 4 = 7) grade cancer called Prolaris, that uses a proprietary panel of gene expression to predict prognosis.

PSMA (GCPII, FOLH1) is an enzyme encoded by folate hydrolase (prostate-specific membrane antigen) 1 (FOLH1). It is expressed in many tissues, including the prostate, kidney, the small intestine, and the central and peripheral nervous system. Unlike PSA, whose expression per cell decreases in cancer compared with benign prostate and which decreases in the highest grades of prostate cancer, the expression of PSMA increases with about a tenfold-higher expression in prostate cancer than benign prostate. PSMA therefore has found use as a diagnostic immunohistochemical stain and PSMA is the target of an approved imaging agent for prostate cancer, capromabpentide, PROSTASCINT (Cytogen Corporation, Princeton, NJ). Various antibodies and ligands are being developed to target PSMA [46, 47].

ERBB2 (HER2), best known for its predictive value in breast cancer, is to a lesser degree prognostic in prostate cancer. 17–20 % of reactive cases are reactive in conventional prostatic adenocarcinoma. Interestingly, however, 65 % percent of prostate cancers with intraductal carcinoma (IDC) are immunoreactive for ERBB2 in the IDC, suggesting that such cases may benefit from targeted Herceptin antibodies [38]. However, because ERBB2 (HER2) reactivity increases with higher Gleason score per se, it is still not known whether the IDC has any greater reactivity than accompanying invasive high-grade acinar components. Another tumor that is in the differential diagnosis for IDC is adenoid cystic/basal cell carcinoma, and that tumor was consistently strongly reactive for ERBB2 [48] so the distinction with IDC must be made morphologically.

The urinary bladder is the most frequent site of cancer in the luminal urinary tract, and the sixth most common cancer in the USA, with an annual incidence of approximately 75,000. According to the SEER factsheet, there are an estimated 571,518 people living in the USA with a diagnosis of bladder cancer, with 15,000 deaths/year [59]. The classical and most common presenting symptom of bladder cancer is painless, intermittent hematuria. Other clinical features include urinary urgency and frequency, incontinence, and dysuria or other obstructive symptoms. Signs suggestive of an invasive malignancy include renal failure, a palpable suprapubic mass, as well as the classical systemic symptoms of weight loss and anorexia.

Tobacco use has been established as the single most important risk factor for the development of bladder cancer. Burch et al. reported that the incidence of transitional cell carcinoma of the bladder is approximately four times higher in smokers than in nonsmokers [60]. Other risk factors include cyclophosphamide use, history of pelvic radiation, and chronic irritation associated with infection such as Schistosomiasis, or the use of long-term indwelling catheters.

Primary urothelial neoplasms of the bladder are broadly categorized as flat or papillary lesions, and further subclassified as invasive or noninvasive. As with epithelial neoplasms in other organs, urothelial carcinomas of the bladder may be low- or high-grade, which may harbor variable molecular and chromosomal aberrancies, and require different diagnostic algorithms. Treatment largely depends on stage and grade; superficial tumors are usually treated with transurethral resection and vesicular chemotherapy/pro-inflammatory agents, while deeply invasive tumors are treated with radical cystectomy and systemic chemotherapy.

Cytoscopy and urine cytology have long been the standard method for monitoring recurrence. Cystoscopy, while highly sensitive, is also an invasive and costly test. Cytology is minimally invasive, but demonstrates a sensitivity of only 44 %, and may be as low as 27 % for low-grade tumors [61].