AR-V7 and efficacy of abiraterone (Abi) and enzalutamide (Enza) in castration-resistant prostate cancer (CRPC): Expanded analysis of the Johns Hopkins cohort.
5012 Background: We previously reported an association between ARV7 detection and poor outcomes with Abi/Enza in 62 CRPC pts (NEJM 2014). Here, we expand our analysis to 202 pts, and also report the prognostic significance of CTC detection in addition to ARV7 detection. Methods: We prospectively enrolled CRPC pts starting Abi or Enza, and examined the prognostic value of CTC detection (+ vs -) and ARV7 detection (+ vs -) using a CTC-based ARV7 mRNA assay. We examined ≥50% PSA responses, PSA progression free survival (PSA-PFS), clinical/radiologic PFS (PFS), and overall survival (OS). We constructed multivariable (MVA) models adjusting for PSA, Gleason sum, number of prior hormone therapies, prior Abi/Enza use, prior taxane use, presence of visceral mets, and ECOG score. We also separately examined the 1st-line (Abi/Enza-naïve) and 2nd-line (post-Abi or Enza) novel hormonal therapy (NHT) settings. Results: 202 pts were enrolled (median f/u 12.9 mo, range 1.3–35.8 mo). CTC+/ARV7+ pts were more likely to have Gleason ≥8 (p= .05), M1 disease at diagnosis (p= .01), higher PSA (p< .01), prior Abi/Enza use (p= .03), prior taxane use (p= .02), and ECOG ≥1 (p= .01). Outcomes for the overall cohort, and separately for the 1st-line and 2nd-line NHT cohorts, are shown in the Table. All correlations remained significant in MVA models. Conclusions: The negative prognostic impact of ARV7 in pts receiving Abi/Enza is confirmed in this expanded analysis, and applies to both the 1st-line and 2nd-line NHT settings. This study is the first to suggest that our CTC-based ARV7 assay should be interpreted using 3 separate prognostic categories (CTC–, CTC+/ARV7-, CTC+/ARV7+).