Androgen deprivation therapy (ADT), either by surgical or medical castration depletes testicular production of testosterone (T) and has long been considered as the mainstay of upfront treatment for advanced prostate cancer. Despite the initial response to this treatment, disease progression to castration-resistant prostate cancer (CRPC) usually occurs. The adrenal production of androgen precursor steroids, such as dehydroepiandrosterone (DHEA) and DHEA-sulfate (DHEA-S), permits intratumoral production of T and 5α-dihydrotestosterone (DHT), activation of androgen receptor (AR) and AR-dependent gene expression. Abiraterone (Abi; given as Abi acetate) and enzalutamide (Enz) are 2 of the agents that target different parts of the androgen pathway by blocking androgen production and signaling, ultimately improving overall survival for patients with metastatic CRPC.1,2 Abi is a potent inhibitor of 17α-hydroxylase/17,20-lyase (CYP17A1), which is a key enzyme required for DHEA and DHEA-S production. Synthesis of DHT, the most potent androgen, from DHEA necessitates a series of enzymes, including 3β-hydroxysteroid dehydrogenase (3βHSD), steroid-5α-reductase (SRD5A), and 17β-hydroxysteroid dehydrogenase (17βHSD) isoenzymes. This process enables the conversion of ADT-responsive prostate cancer into CRPC. Multiple retrospective reviews have shown the limited efficacy of Enz after Abi treatment failure and vice versa.3,4 To date there are very little data on the appropriate sequence of Enz and Abi utilization, which might thereby maximize the potential efficacy and duration of response to treatment with these agents. Our recently published research demonstrates that Δ4, 3-keto-abi (D4A) is a metabolite of Abi that is formed in patients with prostate cancer and has CYP17A1 inhibition activity that is comparable to Abi.5 Furthermore, D4A inhibits 3βHSD (a key enzyme involved in CRPC progression) more potently than Abi and at higher concentrations it demonstrates significant SRD5A inhibitory effects as well. Moreover, although Abi has been shown to have AR antagonistic activity,6 we observed more potent AR inhibition activity with D4A that is comparable to Enz in our experiments.5 As a result, by targeting multiple steps of the AR signaling pathway, D4A showed a significant increase in progression free survival compared to Abi and Enz in our xenograft models, suggesting that it might play an important role in the clinical activity of Abi in CRPC patients. Abi is converted to D4A by 3βHSD. Increased 3βHSD enzymatic function, with the HSD3B1(1245C) genetic variant encoding a missense in 3βHSD1 as one known mechanism, also increases intratumoral flux to DHT that may result in the development of treatment resistance.7 Therefore, D4A conversion ratio could serve as an indicator of 3βHSD activity and ultimately may predict resistance to hormonal therapy for prostate cancer. On the other hand, higher 3βHSD activity would translate into a higher D4A concentration, which is a more potent metabolite than Abi as described above. Although the Abi to D4A conversion ratio is low (at about 5%) in most patients, it appears to be quite variable. A higher conversion ratio might mimic conditions of treatment with Enz, particularly given the AR antagonist activity of D4A. In other words, 3βHSD activity level in a patient might be helpful in predicting the potential response to Abi and ultimately a patient with higher concentration of D4A with Abi treatment might be less likely to benefit from subsequent Enz treatment. Conversely, increasing D4A levels by increasing Abi concentrations or the conversion to D4A might increase the clinical benefit of treatment with Abi. In addition, because 3βHSD is required for mineralocorticoid synthesis, 3βHSD inhibition by D4A might reduce mineralocorticoid production that could translate into a better side effect profile (less hypertension and hypokalemia) and potentially lower dose of prednisone requirement in patients with higher D4A levels. These questions have yet to be answered in clinical studies. Finally, it is possible that steroidogenic metabolites of Abi are not limited to D4A, and that other metabolites with clinically relevant biochemical activity contribute to response and resistance to treatment with Abi.