Introduction

The exact role of pre-treatment (baseline) total serum testosterone (BST) is still controversial in patients with prostate cancer (PCa) and conflicting results are reported in the literature. We assessed the impact of BST on the risk of biochemical failure (BF) in patients with PCa and treated with definitive radiation therapy (RT).

Methods

The current study is a retrospective analysis of 360 prospective patients diagnosed with non-metastatic PCa between 2002 and 2014 and enrolled into seven different prospective multicentric phase II-III trials performed at our institution. All patients received definitive RT after initial diagnostic workup which included PSA and BST assessment. Patients were stratified according to hypogonadal (BST<11 nmol/L) vs. non-hypogonadal state (BST≥11 nmol/L). The ability of this BST cut-off to predict BF was assessed in Kaplan-Meier analyses, as well as in univariable and multivariable Cox regression analyses. Internal validation of our findings was performed using bootstrap resampling with 10,000 replications.

Results

The median (IQR) age at diagnosis was 71 years (65-74). Median (IQR) PSA was 7.7 ng/mL (5.6-12.1). Testosterone ranged from 0.7 to 28.9 nmol/L (mean: 11 nmol/L; median 10.2 nmol/L; IQR 8.3-13.0 nmol/L). The number of patients with BST < 11 nmol/L was 209 (58.0%), while a total of 151 patients had a BST ≥ 11 nmol/L (42.0%). A total of 272 patients (75.6%) had available BMI data, which ranged between 17.5 and 52.6, without significant differences between the two groups (p=0.1). ADT was administered only to 108 patients (30%), including all individuals diagnosed with PSA > 20 ng/mL. Median follow-up was 72 months. Overall, BF-free survival rates at 96 months was 79.6% (95% CI: 72.3-87.6%) in hypogonadal vs. 65.1% (95% CI: 55.3-76.7%) in non-hypogonadal individuals (p=0.042). In multivariable Cox regression analyses, BST < 11 nmol/L was associated with a significantly reduced risk of BF (HR: 0.50; CI 0.30-0.83; p = 0.007). After 10,000 bootstrap resamples, virtually the same results were recorded.

Conclusions

Oncologic outcomes for PCa after primary radiation therapy are affected by pre-treatment testosterone levels. Individuals with lower baseline testosterone levels experienced more favourable biochemical failure rates after adjusting for the use of ADT. As many still debate about the role of testosterone in PCa, our findings need to be validated in larger patient cohorts.

Funding

none