Abstract Background: Prostate cancer studies have used urine as a non-invasive source of nucleic acids for biomarker analysis. These studies have been limited by the need for a digital rectal exam (DRE) or prostate massage prior to urine collection to enable enough cellular material for RNA analysis. Such manipulations can potentially lead to variability in assay performance depending on DRE intensity and may also reduce compliance. Here we investigate the novel use of RNA in urinary exosomes, small lipid bilayer vesicles released from cells into bodily fluids, to analyze previously identified prostate cancer mRNA biomarkers without a prostate massage. Methods: Random urine samples (>20ml) were collected from 207 consecutive patients under a Columbia University IRB approved prospective protocol. Patients were stratified into 5 groups: TRUS biopsy negative (Bx Neg, n=39), TRUS biopsy positive (Bx Pos, n=47), post-radical prostatectomy (RP, n=37), no TRUS biopsy (No Bx, n=44) and control (healthy males <35 yrs, n=40). Due to the unique stability of exosomal RNA (exoRNA), urine samples were stored at 4°C and 0.8 um filtration was used to remove whole cells and debris. Urinary exoRNA was isolated using our in-house technique. RT-qPCR was used to analyze GAPDH, prostate-specific antigen (PSA), prostate cancer antigen 3 (PCA3), androgen receptor (AR), survivin, nuclear receptor co-activator 2 (NCOA2), RAD21, transmembrane protease serine 2 (TMPRSS2), ERG and TMPRSS2:ERG fusion. Results: All groups were age matched except for control males (<35 yrs)(P<0.001). Serum PSA protein was significantly lower in the No Bx and RP groups versus the Bx Neg and Bx Pos groups (P<0.001). Relative quantitation (RQ) of genes standardized to GAPDH revealed that ERG and PCA3 expression levels were significantly increased in the Bx Pos versus Bx Neg groups (ERG P<0.01, PCA3 P<0.05). Further, RQ analysis standardized to the prostate related gene, PSA, revealed that survivin (P<0.005), ERG (P<0.005), PCA3 (P<0.005), TMPRSS2:ERG fusion (P<0.05), and TMPRSS2 (P<0.05) were significantly increased in the Bx Pos versus Bx Neg group. Analysis of the TMPRSS2:ERG fusion revealed 44% of Bx Neg and 68% of Bx Pos had a positive fusion event. Analysis of the RP group revealed that no patients had positive TMPRSS2:ERG detection; this finding was further supported by the loss of TMPRSS2:ERG expression experienced by 4 Bx Pos patients following prostatectomy suggesting specificity of the fusion event for the prostate. Interestingly, the control group (males <35yrs) had a much lower TMPRSS2:ERG fusion incidence of 5% compared to the No Bx group's fusion incidence of 41%, which was similar to the Bx Neg group despite having significantly lower serum PSA protein levels. Conclusion: This study confirms that urinary exosomes are a novel source of high quality RNA to examine prostate gene expression. We demonstrate that expression levels of ERG, PCA3, TMPRSS2:ERG, TMPRSS2 and survivin are significantly higher in cases of confirmed prostate cancer, consistent with previous studies using prostate biopsies and/or urinary cells collected after DRE/prostate massage. The unique stability and yield of urinary exoRNA collected without the need for a prostatic massage will hopefully simplify sample handing, obviate sample variability and patient discomfort inherent to prostate massage, and broaden the role of exosomes in future diagnostic testing. Citation Format: {Authors}. {Abstract title} [abstract]. In: Proceedings of the AACR-NCI-EORTC International Conference: Molecular Targets and Cancer Therapeutics; 2011 Nov 12-16; San Francisco, CA. Philadelphia (PA): AACR; Mol Cancer Ther 2011;10(11 Suppl):Abstract nr C133.