Alternative splicing of the gene encoding the androgen receptor (AR) has been proposed as a mechanism of therapeutic resistance to AR signalling (ARS) inhibitors. The splice variant AR-V7 lacks the ligand-binding domain, leading to constitutive activation of ARS. Howard Scher and colleagues have now validated nuclear AR-V7 in circulating tumour cells (CTCs) as a marker to guide treatment decisions for patients with metastatic castration-resistant prostate cancer (mCRPC).

Blood samples were obtained from patients with mCRPC before starting treatment with ARS inhibitors (n = 70) or with a taxane (n = 72). Patients in which ≥1 CTC with an intact nucleus and nuclear staining for AR-V7 was detected in ~1 ml of blood were defined as AR-V7+. Overall survival data were analysed separately for patients with high-risk disease (n = 70) or low-risk disease (n = 72). Patients with AR-V7− high-risk disease had a longer median overall survival with ARS inhibitors than with taxanes (16.9 months versus 9.7 months; HR 2.38; P = 0.02), whereas patients with AR-V7+ high-risk disease had a longer median overall survival with taxanes than with ARS inhibitors (14.3 months versus 5.6 months; HR 0.35; P = 0.03). In the low-risk group, the number of AR-V7+ patients was not sufficient to estimate differences in survival between treatments according to the presence of AR-V7. Patients with low-risk disease should receive taxanes if they are AR-V7+ or ARS inhibitors if they are AR-V7–.

The CTC detection test used in this study is commercially available; Scher hopes that it is implemented in routine clinical practice. “Our focus was to show that the use of the biomarker test result to inform the choice of treatment could lead to an improved outcome for patients relative to not using the test result,” Scher comments, adding “AR-V7 status alone does not explain all of the mechanisms of resistance to AR inhibitors. Going forward, we will take the approach used in this study to identify and validate other markers of resistance, including tumour heterogeneity and genomic instability.”