Beyond castration: defining maximal testosterone control in advanced prostate cancer.

Objective: This study aimed to investigate the correlation between the minimum testosterone (T) level achieved during androgen deprivation therapy (ADT) for advanced prostate cancer and progression and prognosis. And to establish the new recommended threshold for defining castration-level testosterone. Methods: A retrospective analysis was conducted on 425 patients with advanced prostate cancer undergoing ADT. Patients were stratified into three groups based on their lowest testosterone level: castration low<10 ng/dL, castration 10-50 ng/dL, Non-castrated >50 ng/dL. To further explore subgroup progression and survival differences in low castrated testosterone levels, those castrated low testosterone levels were divided into two groups, castration ultra-low 5-10 ng/dL and castration extreme low<5ng/dL. Additionally, a small cohort (N = 29) of surgically castrated patients was included for subgroup analysis. Correlations between the minimum testosterone level and outcomes, time to progression (TTP) and overall survival (OS). Results: Significant differences in TTP were observed among the three groups (P<0.001), and both two groups (P<0.001). The castration low T level group had TTP of 24.62 ± 13.62 months and the lowest percentage of TTP<18 months (33.88%), the castration T level group had TTP of 15.65 ± 9.16 months with the second highest percentage of TTP<18 months (64.34%), the non-castrated T level group had TTP of 10.93 ± 7.89 months with the highest percentage of TTP<18 months (83.33%). There was a significant difference in survival rates between the three groups (P<0.001). Differences were found between the both two groups (P<0.01), with the castration low T level group demonstrating superior 3- and 5-year survival rates compared to the other groups. The non-castrated T level group had the worst prognosis. No significant differences in TTP or survival rates were observed between the castration ultra-low and extreme-low T subgroups. However, surgically castrated patients exhibited the poorest prognosis. Minimum testosterone level was weakly negatively correlated with TTP (r = -0.32, P< 0.001), but not significantly correlated with OS. Conclusion: Challenging the traditional castration standard, this study identifies 10 ng/dL (versus 50 ng/dL) as the critical testosterone threshold for evaluating tumor progression and prognosis in advanced prostate cancer patients on ADT.