A Systematic Review and Meta-Analysis on the Predictive Value of Cell-Free DNA-Based Androgen Receptor Copy Number Gain in Patients With Castration-Resistant Prostate Cancer

JCO Precis Oncol. 2020 Nov:4:714-729. doi: 10.1200/PO.20.00084.

Abstract

Purpose: It has been suggested that androgen receptor copy number gain (AR gain) detected in cell-free DNA (cfDNA) can predict treatment response to androgen receptor signaling inhibitors (ARSIs) in patients with castration-resistant prostate cancer (CRPC). But it is unclear whether cfDNA-based AR gain is a true resistance mechanism to ARSIs or mainly a reflection of the tumor burden. In this systematic review, we aim to summarize current literature and comment on the potential of cfDNA-based AR gain as a predictive biomarker to guide therapy choices.

Methods: A literature search was conducted in PubMed/Medline, Cochrane, Embase, and Web of Science databases. Sixteen articles published before November 2019 were selected for the meta-analysis, representing more than 1,000 patients. By using a random effects model, the progression-free survival (PFS) and overall survival (OS) were compared between patients with and without cfDNA-based AR gain who had been treated with ARSIs or with taxane chemotherapy.

Results: Upon treatment with ARSIs, the PFS (hazard ratio [HR], 2.33; 95% CI, 2.00 to 2.72; P < .0001) and the OS (HR, 3.83; 95% CI, 3.11 to 4.70; P < .0001) were worse for patients with cfDNA-based AR gain, independent of the line and type of ARSIs. The OS and PFS in patients treated with first-line docetaxel or second-line or third-line cabazitaxel seemed to be unaffected by AR gain, despite a higher disease burden in patients with AR gain. AR gain was associated with reduced response with later lines of docetaxel.

Conclusion: In patients with CRPC, cfDNA-based AR gain is associated with a worse response to ARSIs. The effect on patients who are receiving taxane chemotherapy seems to be dependent on the type and line, although data are limited. Future prospective studies are essential to assess the true potential of cfDNA-based AR gain as a minimally invasive biomarker to guide therapy choice.