Abstract Localized intermediate and high-risk prostate cancer can be treated with androgen deprivation therapy (ADT). Initially, patients undergo remission but inevitably relapse due to the emergence of castration resistant prostate cancer (CRPC). Newer agents such as the P450c17 inhibitor, abiraterone acetate (AA), have gained approval for the treatment of CRPC and increased median survival by 4 months. We have developed a novel and validated stable isotope dilution liquid chromatography selected reaction monitoring mass spectrometry (SID-LC/ESI/SRM/MS) method for the quantification of conjugated and unconjugated keto-androgens in human serum (J Steroid Biochem Mol Biol (2013) 138:281). This method was applied to human serum from patients enrolled in a neoadjuvant AA clinical trial (J Clin Oncol (2014) 32:3705). We found that testosterone (T) and 5α-dihydrotestosterone (DHT) did not always decrease in tandem, which suggests that pathways that bypass T may lead to DHT. These may include the alternative pathway (Δ4-androstene-3,17-dione -> 5α-androstane-3,17-dione -> DHT) or the backdoor pathway (androsterone -> 5α-androstane-3α,17β-diol -> DHT). Second, despite achieving >90% inhibition of P450c17, the level of dehydroepiandrosterone sulfate (DHEA-S) that remains (20,000 ng/dL) is 4,000-fold higher than castrate levels of T achieved in the trial. We hypothesize that the DHEA-S depot that remains is sufficient to feed intratumoral androgen biosynthesis and that this could account for the clinical failure of P450c17 inhibitors in CRPC. Adaptive intratumoral androgen biosynthesis may be facilitated by the up-regulation of AKR1C3 (type 5 17β-hydroxysteroid dehydrogenase). In order to test our hypothesis, we have developed a panel of isogenic prostate cancer cell lines that either express AKR1C3 or are AKR1C3 null. These cell lines are then challenged with post-AA levels of circulating androgens (DHEA-S, DHEA and Δ4-androstene-3,17-dione) to determine whether they can still make sufficient androgens to activate the androgen receptor (AR), where androgen metabolism is measured using SID-LC/ESI/SRM/MS. Using LNCaP and LNCaP-AKR1C3 cells, we find that the level of T detected as T-17β-glucuronide in the AKR1C3 expressing cells could be sufficient to activate the AR. These results would suggest that AKR1C3 inhibitors in combination with AA could benefit patients who might otherwise fail AA treatment. [Supported by grants: R25-CA101871 to DT; P30-ES-13508 to TMP; P01-CA-163337 to SPB, EM, TMP and PSN)] Citation Format: Daniel Tamae, Elahe Mostaghel, Bruce Montgomery, Peter S. Nelson, Steven P. Balk, Philip W. Kantoff, Mary-Ellen Taplin, Trevor M. Penning. Resistance to P450c17 inhibitors in castration-resistant prostate cancer may result from the DHEA-S depot that remains and can be used by AKR1C3 for intratumoral androgen biosynthesis. [abstract]. In: Proceedings of the 106th Annual Meeting of the American Association for Cancer Research; 2015 Apr 18-22; Philadelphia, PA. Philadelphia (PA): AACR; Cancer Res 2015;75(15 Suppl):Abstract nr 3450. doi:10.1158/1538-7445.AM2015-3450