Androgen deprivation therapy (ADT) is the standard of care for prostate cancer treatment. Studies suggest that patients with testosterone levels below 0.7 nM have a longer time to castration resistance. Using the most accurate testosterone measurement method, namely mass spectrometry (MS), we sought to determine if a lower testosterone level under ADT could be associated with longer time to castration resistance.

This retrospective study included 138 prostate cancer patients undergoing noncurative continuous ADT for which we had access to testosterone measurements assessed by MS. For 108 samples, paired immunoassays (IA) testosterone measurement was available. Primary outcome was time to castration-resistant prostate cancer (CRPC). The Contal and O'Quigley method was used to determine the optimal testosterone castration cut-off point considering the outcome and time-to-event variables. Relationship between testosterone levels assessed either by IA or MS and time to CRPC was evaluated using Cox regression.

Mean testosterone level was 0.370 nM by IA and 0.275 nM as assessed by MS. The optimal testosterone cut-off point identified to predict time to CRPC was of 0.705 nM for IA and of 0.270 nM for MS. While no significant difference for time to CRPC was found between patients showing IA testosterone level ≥0.705 nM versus <0.705 nM (hazard ratio [HR]: 1.579; 95% confidence interval [CI]: 0.908-2.745), patients with MS testosterone ≥0.270 nM had an increased risk of progression to CRPC compared to MS testosterone <0.270 nM in univariate (HR: 1.717; 95% CI: 1.160-2.541) and multivariate analysis (HR: 1.662; 95% CI: 1.043-2.648).

The higher sensitivity of MS testosterone measurement methods allows the identification of a lower castration threshold and leads to early identification of patients more likely to progress to CRPC. These patients would likely benefit from treatment intensification by androgen receptor axis-targeted therapies to delay disease progression.