Tumor-derived exosomes, which are nanometer-sized extracellular vesicles of endosomal origin, have emerged as promoters of tumor immune evasion but their role in prostate cancer (PC) progression is poorly understood. In this study, we investigated the ability of prostate tumor-derived exosomes to downregulate NKG2D expression on natural killer (NK) and CD8+ T cells. NKG2D is an activating cytotoxicity receptor whose aberrant loss in cancer plays an important role in immune suppression. Using flow cytometry, we found that exosomes produced by human PC cells express ligands for NKG2D on their surface. The NKG2D ligand-expressing prostate tumor-derived exosomes selectively induced downregulation of NKG2D on NK and CD8+ T cells in a dose-dependent manner, leading to impaired cytotoxic function in vitro. Consistent with these findings, patients with castration-resistant PC (CRPC) showed a significant decrease in surface NKG2D expression on circulating NK and CD8+ T cells compared to healthy individuals. Tumor-derived exosomes are likely involved in this NKG2D downregulation, since incubation of healthy lymphocytes with exosomes isolated from serum or plasma of CRPC patients triggered downregulation of NKG2D expression in effector lymphocytes. These data suggest prostate tumor-derived exosomes as down-regulators of the NKG2D-mediated cytotoxic response in PC patients, thus promoting immune suppression and tumor escape.

Background Novel therapies for castration resistant prostate cancer (CRPC) have been introduced in the clinic with possibilities for individualized treatment plans. Best practice of those expensive drugs requires predictive biomarker monitoring. This study used circulating biomarker analysis to follow cancer‐derived transcripts implicated in therapy resistance. Method The isolated platelet population of blood samples and digital‐PCR were used to identify selected biomarker transcripts in patients with CRPC prior chemo‐ or androgen synthesis inhibiting therapy. Results Fifty patients received either docetaxel (n = 24) or abiraterone (n = 26) therapy, with therapy response rates of 54% and 48%, respectively. Transcripts for the PC‐associated biomarkers kallikrein‐related peptidase‐2 and ‐3 (KLK2, KLK3), folate hydrolase 1 (FOLH1), and neuropeptide‐Y (NPY) were uniquely present within the platelet fraction of cancer patients and not detected in healthy controls (n = 15). In the abiraterone treated cohort, the biomarkers provided information on therapy outcome, demonstrating an association between detectable biomarkers and short progression free survival (PFS) (FOLH1, P < 0.01; KLK3, P < 0.05; and NPY, P < 0.05). Patients with biomarker‐negative platelets had the best outcome, while FOLH1 (P < 0.05) and NPY (P = 0.05) biomarkers provided independent predictive information in a multivariate analysis regarding PFS. KLK2 (P < 0.01), KLK3 (P < 0.001), and FOLH1 (P < 0.05) biomarkers were associated with short overall survival (OS). Combining three biomarkers in a panel (KLK3, FOLH1, and NPY) made it possible to separate long‐term responders from short‐term responders with 87% sensitivity and 82% specificity. Conclusion Analyzing tumor‐derived biomarkers in platelets of CRPC patients enabled prediction of the outcome after abiraterone therapy with higher accuracy than baseline serum PSA or PSA response.

e583 Background: Novel therapies for castration resistant prostate cancer (CRPC) have been introduced in the clinic with possibilities for individualized treatment plans. Best practice of those expensive drugs requires predictive biomarker monitoring. This study used circulating biomarker analysis to follow cancer-derived transcripts implicated in therapy resistance. Methods: The isolated platelet population of blood samples and QRT-PCR were used to identify selected biomarker transcripts in patients with CRPC prior chemo- or androgen synthesis directed therapies. The association between biomarker status in platelets (positive vs. negative) and therapy response, progression-free survival (PFS) and overall survival (OS) was examined. Results: Fifty patients received either docetaxel (n = 24) or abiraterone (n = 26) therapy, with therapy response rates of 54% and 48%, respectively. Transcripts for the PC-associated biomarkers kallikrein-related peptidase-3 ( KLK3), androgen receptor splice-variant 7 (ARV7), folate hydrolase 1 (FOLH1), and neuropeptide-Y ( NPY) were present within the platelet fraction. Analyzing biomarkers in the chemotherapy group did not add information about PFS, but FOLH1 was associated with short OS (p = 0.00015). In the abiraterone treated cohort, detectable FOLH1 (p = 0.009), KLK3 (p = 0.018), and NPY (p = 0.028) were all associated with short PFS. Patients with biomarker-negative platelets had the best outcome, while FOLH1 and NPY provided independent predictive information regarding PFS and KLK3 (p = 0.001) and FOLH1 (p = 0.002) were associated with short OS. Conclusions: Analyzing tumor-derived biomarkers in platelets of CRPC patients enabled prediction of the outcome after abiraterone therapy with high accuracy. Platelet-based analysis of FOLH1, NPY, and KLK3 may be used for treatment stratification of patients scheduled for treatment with abiraterone.