cers produced significantly (p=0.049) more IL-8 (median: 15.4; IQR: 6.2-24.8) than cultures derived from patients har- bouring less aggressive cancers (median: 6.9; IQR: 3.8-9.9). The production of IL-8 induced by poly(I:C) did not differ across cancer groups (p=0.182). Use of NSAID before surgery did not differ (Fisher’s ex- act p=0.259) between patients with more or less aggressive cancers; and did not affect IL-8 production. Conclusions: A high baseline pro-inflam- matory status of normal prostate epithe- lium is associated with a more aggressive form of prostate cancer. More research to decipher the possibly causal relationship linking inflammation to prostate carcino- genesis is needed. UP-02.085 Antigen-Specific T-Cell Responses Against Tumor-Antigens Are Controlled by Regulatory T-Cells in Prostate Cancer Patients Hadaschik B 1 , Su Y 1 , Hadaschik E 2 , Hohenfellner M 1 , Beckhove P 3 Depts. of 1 Urology, 2 Dermatology, University Hospital Heidelberg, 3 Translational Immunology Unit, German Cancer Research Center, Heidelberg, Germany Introduction and Objective: Immuno- therapeutic approaches belong to the most promising regimen in an effort to control castration-resistant prostate cancer (CRPC). The objective of this study was to characterize the natural repertoire of tu- mor-antigen-reactive T-cells in patients with prostate cancer and to analyze the suppression of anti-tumor responses by regulatory T-cells (Treg). Materials and Methods: After written informed consent, peripheral blood sam- ples were collected from 57 patients with histologically confirmed prostate cancer, 8 patients with BPH, and 16 healthy donors. Peripheral blood mononuclear cells were isolated and antigen-specific IFN- secre- tion of isolated T-cells was analyzed by ELISPOT assays. T-cells were functionally characterized and T-cell responses before and after Treg-depletion were compared. As test tumor-antigens, a panel of 11 long synthetic peptides derived from 8 differ- ent tumor-antigens including PSA and PAP was used. Results: In prostate cancer patients, an effector T-cell response rate of 74.5% compared with only 25% in BPH and 31% in healthy donors was observed. Most pa- tients recognized 2-3 different tumor-anti- gens. Looking at various stages of the dis- ease, there was a clear increase in immune responses against prostate-spe- cific antigens from intermediate to high risk tumors and castration-resistant dis- ease. Depletion of Tregs led to a signifi- cant boost in effector T-cell responses against PSA and PAP. Conclusions: Tumor-specific effector T- cells were detectable in the majority of prostate cancer patients and especially in CRPC. Since effector T-cell responses against prostate-specific antigens strongly increased after Treg depletion, our results indicate that the efficacy of immunotherapy could be enhanced by reduction of Tregs. UP-02.086 Comprehensive Analysis of Intratumoral Steroidogenesis in Castration Resistance Prostate Cancer Hamid A 1 , Aalders T 1 , Jansen C 1 , Verhaegh G 1 , Pfeiffer M 1 , Smit F 2 , Schalken J 1 1 Dept. of Urology, Radboud University Nijmegen Medical Centre, 2 NovioGendix, Nijmegen, Netherlands Introduction and Objective: Molecular mechanisms associated with Castration Resistance Prostate Cancer (CRPC) have gained considerable attention in recent years. It has been hypothesized that CRPC still depends on the androgen signaling axis. Our microarray data showed that the AKR1C3 and Androgen Receptor (AR) genes were the only ones significantly up regulated in CRPC. Several studies showed that several AR splice variants (ARV) could also be important in progression to CRPC. The aim of this study is to evaluate comprehensively the expression of the full length AR (ARfl), AR splice variant and AKR1C3 genes involved in androgen signaling axis in CRPC specimens. Materials and Methods: We used TURP tissues from CRPC patients and radical prostatectomy tissues for primary prostate cancer (PrCa). We analyzed the expres- sion of AKR1C3, ARfl and several ARVs (AR1/2/2b, AR3, ARV1, ARV7 and ARV567es) by RT-qPCR and/or Western blotting. In addition, we analyzed muta- tions in the AR ligand binding domain by DNA sequence analysis. We classified CRPC samples into four groups: high AKR1C3 and high ARfl expression (group 1); low AKR1C3 and high ARfl expression (group 2); high AKR1C3 and low ARfl expression (group 3) and low AKR1C3 and low ARfl expression (group 4). Results: We analyzed 17 CRPC and 20 primary PrCa tissues. Expression of AKR1C3 and ARfl mRNA were significantly up regu- lated in 65% and 77% of CRPC samples, re- spectively. However increased expression of AKR1C3 protein was only seen 30% of CRPC patients. In CRPC, ARV were ex- pressed at only 1% of ARfl levels, whilst ARV were hardly expressed in primary PrCa. AR1/2/2b levels were significantly correlated with ARfl, whilst AR3 and ARV7 were not. However, in contrast to ARfl, AR3 and ARV7 were significantly correlated to the levels of AKR1C3 mRNA. There are only 2 patients (12%) had both low AKR1C3 and ARfl expression. Conclusion: In the majority of CRPC tissues, steroidogenic enzyme AKR1C3 or AR receptor expression levels are in- creased. These data suggest that the ma- jority of CRPC tumors may still depend on AR and AR signaling to survive and prolif- erate. Due to heterogeneity in AKR1C3 and AR expression, CRPC patients may be classified for individualized treatment. UP-02.087 Gastrin-Releasing Peptide: Novel Predictor of Castration-Resistant Prostate Cancer? Heinrich E, Probst K, Michel M, Trojan L Dept. of Urology, University Medical Center Mannheim, Mannheim, Germany Introduction and Objective: Neuroen- docrine (NE) cells of the prostate are known to be androgen independent and NE peptides like gastrin-releasing peptide (GRP) or neuron-specific enolase (NSE) can stimulate growth in a paracrine man- ner, and this is thought to be one of the escape mechanisms in castration-resistant prostate cancer (CRPCa). In a longitudinal study, we investigated the development of the NE serum factors GRP, NSE, and Chro- mogranin A and their correlation with prostate-specific androgen (PSA) during hormonal treatment. Materials and Methods: Thirty-two pa- tients, with histology-proven, localized or metastatic prostatic carcinoma (PCa), who were undergoing therapy with LHRH ana- logue or a combination of LHRH analog and peripheral androgen blockade, took part in the study. In addition, eight healthy volunteers were each tested twice for serum GRP to elicit a “physiological” standard value. Blood samples were taken periodically from each patient within an 18-month time frame. Results: We defined the standard value for GRP in the healthy participants as 0.852 ng/ml (mean+2SD) and observed that the GRP values for patients with PCa were significantly higher (P=0.034). There was a positive correlation between UNMODERATED POSTER SESSIONS UROLOGY 78 (Supplement 3A), September 2011 S289