BACKGROUND AND OBJECTIVEThe PRESIDE (NCT02288247) randomized trial demonstrated prolonged progression-free survival (PFS) with continuing enzalutamide beyond progression in metastatic castration-resistant prostate cancer (mCRPC) patients starting docetaxel. This study aims to test the associations of PFS and circulating tumor DNA (ctDNA) prior to and after one cycle (cycle 2 day 1 [C2D1]) of docetaxel and with a liquid biopsy resistance biomarker (LBRB; plasma androgen receptor [AR] gain and/or circulating tumor cells [CTCs] expressing AR splice variant 7 [CTC-AR-V7]) prior to continuation of enzalutamide/placebo.METHODSPatients consenting to the biomarker substudy and donating blood before starting docetaxel with enzalutamide/placebo (N = 157) were included. Sequential plasma DNA samples were characterized with a prostate-cancer bespoke next-generation-sequencing capture panel (PCF_SELECT), and CTCs were assessed for AR-V7 (Epic Sciences, San Diego, CA, USA). Cox models, Kaplan-Meier, and restricted mean survival time (RMST) at 18 mo were calculated.KEY FINDINGS AND LIMITATIONSThere was a significant association of worse PFS with pre-docetaxel ctDNA detection (N = 86 (55%), 8.1 vs 10.8 mo hazard ratio [HR] = 1.78, p = 0.004) or persistence/rise of ctDNA at C2D1 (N = 35/134, 5.5 vs 10.9 mo, HR = 1.95, 95% confidence interval [CI] = 1.15-3.30, p = 0.019). LBRB-positive patients (N = 62) had no benefit from continuing enzalutamide with docetaxel (HR = 0.78, 95% CI = 0.41-1.48, p = 0.44; RMST: 7.9 vs 7.1 mo, p = 0.50). Conversely, resistance biomarker-negative patients (N = 87) had significantly prolonged PFS (HR = 0.49, 95% CI = 0.29-0.82, p = 0.006; RMST: 11.5 vs 8.9 mo, p = 0.005). Eight patients were unevaluable. An exploratory analysis identified increased copy-number gains (CDK6/CDK4) at progression on docetaxel. Limitations included relatively low detection of CTC-AR-V7. Validation of impact on overall survival is required.CONCLUSIONS AND CLINICAL IMPLICATIONSLiquid biopsy gives an early indication of docetaxel futility, could guide patient selection for continuing enzalutamide, and identifies cell cycle gene alterations as a potential cause of docetaxel resistance in mCRPC.PATIENT SUMMARYIn the PRESIDE biomarker study, we found that detecting circulating tumor DNA in plasma after starting treatment with docetaxel (chemotherapy) for metastatic prostate cancer resistant to androgen deprivation therapy can predict early how long patients will take to respond to treatment. Patients negative for a liquid biopsy resistance biomarker (based on the status of androgen receptor (AR) gene and AR splice variant 7 in circulating tumor cells) benefit from continuing enzalutamide in combination with docetaxel, while patients positive for the resistance biomarker did not. Additionally, we identified alterations in the cell cycle genes CDK6 and CDK4 as a potential genetic cause of resistance to docetaxel, which may support testing of specific drugs targeting these alterations.