NATURE REVIEWS | UROLOGY ADVANCE ONLINE PUBLICATION | 1 Introduction The androgen receptor (AR) is an impor- tant therapeutic target in the treatment of metastatic prostate cancer, as evidenced by the life-prolonging effects of androgen deprivation therapy (ADT). Despite the initial efficacy of ADT, tumour regrowth will eventually occur in nearly all patients, resulting in so-called castration-resistant prostate cancer (CRPC). For decades, the cytotoxic agent docetaxel has been the only known life-prolonging drug for patients with CRPC. Of note, docetaxel also affects androgen signalling, as it interferes with the nuclear translocation of the AR. 1 Surprisingly, the AR remains a driver of tumour growth and a useful target in the castration-resistant and even the docetaxel- resistant setting as proven by the effi- cacy of second generation AR-targeting therapeutics, such as enzalutamide and abiraterone. Abiraterone acetate prevents testicu- lar, adrenal and intratumoural androgen synthesis resulting in a survival benefit both before and after chemotherapeutic treatment. 2 However, a major drawback of this drug is the need for simultaneous treat- ment with prednisone, mainly to prevent excessive production of mineralocorticoids and the concomitant adverse effects. Another big step forward in this field was the introduction of enzalutamide, which is a second generation AR antagonist with a higher affinity to the AR compared with the first generation antiandrogens bicalu- tamide and hydroxyflutamide. In 2012, the AFFIRM study demonstrated an average increase in overall survival of 4.8 months in the enzalutamide versus the placebo group in patients with CRPC. 3 Unfortunately, although this new drug has only been available in clinical practice for 2 years, development of resistance has already been noted in the majority of patients. 3 Clearly, defining the unde rlying mechanisms of resistance should lead to the optimization of the treatment, as well as the identifica- tion of new therapeutic targets and pos- sible cross-resistances of prostate tumours to other treatments. In this Perspectives article, we provide a critical review of the recently reported mechanisms of enzalu- tamide resistance in preclinical models of prostate cancer and the clinical setting. Enzalutamide resistance ADT and treatment with AR antagonists have the same goal: lowering AR signal- ling either indirectly by depriving the receptor of its ligand or directly by block- ing receptor activity. Frequently observed mechanisms of resistance to ADT and first-generation antagonists involve the AR. AR gene amplification and resulting increased protein levels, receptor muta- tion and truncation, intracrine synthesis of androgens by the tumour cells, changes in AR cofactor balance, as well as multiple receptor-unrelated mechanisms have been described (Figure 1). 4 Some of these mecha- nisms have also been identified during enzalutamide treatment and various other new mechanisms have been suggested. AR truncations The AR consists of a central DNA- binding domain (DBD), a carboxytermi- nal ligand-binding domain (LBD) and an aminoterminal activation domain (NTD) (Figure 2). 5 In 2009, AR variants that lack the LBD were reported in patients with advanced prostate cancer. 6,7 Loss of the LBD was caused by somatic mutations that introduce stop codons in the AR gene by intragenic deletions or by alternative spli- cing of the primary transcript. 8 Alternative splicing is a naturally occurring mechanism that can generate functionally diverse pro- teins from the same gene. In cancer cells, this mechanism can increase the variations in oncogenic as well as tumour suppressor proteins. However, truncated ARs can also arise through somatic mutations that intro- duce stop codons or through rearrange- ments in the AR gene. Transgenic mice that overexpress truncated ARs (alongside nonmutated variants) in prostate epithelia develop adenomas, demonstrating that the mutant receptor can aid carcinogenesis. 9 Structure–function analyses of the AR protein revealed that truncations of the LBD result in a highly active, ligand- independent receptor mutant. Not surpris- ingly, these truncated forms are unaffected by AR agonists or antagonists since they lack the LBD. 7,10 Moreover, the genes that are induced by truncated ARs are very similar to those of a full-length receptor that is acti- vated by levels of androgens that support OPINION Emerging mechanisms of enzalutamide resistance in prostate cancer Frank Claessens, Christine Helsen, Stefan Prekovic, Thomas Van den Broeck, Lien Spans, Hendrik Van Poppel and Steven Joniau Abstract | The majority of prostate cancers are hormone-dependent at diagnosis highlighting the central role of androgen signalling in this disease. Surprisingly, most forms of castration-resistant prostate cancer (CRPC) are still dependent on the androgen receptor (AR) for survival. Therefore, the advent of new AR-targeting drugs, such as enzalutamide, is certainly beneficial for the many patients with metastatic CRPC. Indeed, this compound provides a substantial survival benefit—but it is not curative. This Perspectives article describes the different ways through which cancer cells can become resistant to enzalutamide, such as AR truncation and other mutations, as well as by-pass of the AR dependence of prostate cancer cells through expression of the glucocorticoid receptor. The clinical relevance of these mechanisms and emerging questions concerning new therapeutic regimens in the treatment of metastatic CRPC are being discussed. Claessens, F. et al. Nat. Rev. Urol. advance online publication 16 September 2014; doi:10.1038/nrurol.2014.243 Competing interests The authors declare no competing interests. PERSPECTIVES © 2014 Macmillan Publishers Limited. All rights reserved