The use of multiparametric Magnetic Resonance Imaging (mpMRI) to identify and site significant prostate tumours is emerging as a promising new diagnostic tool [9], but the high cost of mpMRI and the fact that it misses up to 10% of significant tumours, means the need for inexpensive, non-invasive diagnostic and prognostic tests to help inform patient treatment course remains. An accurate early detection test will reduce costs and morbidity associated with unnecessary invasive procedures and promises to improve outcomes for those with high risk disease.

Blood is considered the most desirable sources of cancer biomarkers because of the socially acceptable and routine nature of collection and processing. Both whole cell and molecular approaches are in fact showing promise for detection of disseminated prostate cancer and assessment of treatment response. Tumour cells enter the circulation during the process of metastasis, but these circulating tumour cells (CTCs) are very rare initially, with only one in a billion circulating cells being of tumour origin in patients with known metastatic disease [10]. The number of CTCs increases with tumour burden in these patients, which has enabled development of an FDA approved CTC enumeration test, CellSearch® to assess the presence of metastatic prostate cancer [11]. However, CTC enumeration is of no predictive value for localised prostate cancer and is not useful as an early detection tool [12]. Molecular (i.e. protein, RNA, microRNA, DNA) and metabolic serum markers are expected to provide greater sensitivity and specificity than whole cell approaches. Some of the most exciting, new molecular findings for prostate cancer involve the use of microRNA panels to detect recurrence of advanced disease following radical prostatectomy [13] and as prognostic markers for late stage disease [14, 15].  While blood markers clearly hold much promise for assessment at this phase, accurate early detection and prognosis is yet to be realised. Higher concentrations of molecular and metabolic biomarkers that provide a more representative portrait of pre-metastatic, localised disease may be obtained from a body fluid proximal to the tumour and into which effluent from the prostate is normally discharged.

Urine is widely accepted as a more specific, prostate representative body fluid than serum. However, only 15% of freely-voided urine samples contain PSA [16], meaning that prostate cells and secretions do not shed into the urethra at a rate high enough to facilitate clinical urine analyses. Vigorous DRE and prostatic massage is required to dislodge prostate cells along with prostatic secretions into the urethra for expulsion during urination [16]. From here, tumour-derived epithelial cells and prostatic secretions can be harvested for various molecular analyses. A non-coding RNA, Prostate Cancer Antigen 3 (PCA3, initially designated DD3), became a target for such analyses following the discovery that it is strongly and specifically up-regulated in over 90% of prostate cancer tissue samples [17].  Detection of PCA3 in the epithelial cell fraction of DRE urine is the principle behind the FDA-approved Progensa® test for prostate cancer [18, 19]. This test has a higher specificity (i.e. does not produce as many false positives) but lower sensitivity (i.e. produces more false negatives) compared with serum PSA and as such, may be beneficial as an adjunct to serum PSA but never as a stand-alone diagnostic test [20]. More recently, a panel of four microRNAs obtained from DRE/prostatic massage secretions was shown to have significantly higher predictive power for prostate cancer than serum PSA [21]. While DRE urine is certainly a source of useful biomarkers, the collection method is inherently biased to the posterior aspect of the prostate so that anterior tumours are likely to remain undetectable using this body fluid. 

In contrast to DRE urine, seminal fluid (SF) is released following vigorous and global prostatic smooth muscle contractions during ejaculation. The potential utility of SF for early diagnosis of prostate cancer was first highlighted by the finding that abnormal prostate cells are detectable in this body fluid well before pathological change is evident on biopsy [22]. Recently, RNA sequencing of the non-sperm epithelial cell fraction of SF uncovered microRNA markers that exhibit higher predictive value for prostate cancer than serum PSA [23]. A follow up study has shown that combining a number of RNA markers from SF with serum PSA further improves predictive value [24]. Reports have also shown that SF is a source of peptide [25], protein [26] and metabolite markers [27] of prostate cancer.  This should not be surprising given that prostate-derived material constitutes ~40% of total SF [28], making it the obvious body fluid for isolation and purification of prostate-associated proteins [29, 30]. SF is also a rich source of prostate-derived extracellular vesicles or ‘prostasomes’ containing cell-free RNA and proteins [31, 32]: a treasure trove of potential prostate cancer biomarkers.