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.