Baseline endogenous serum testosterone levels and all-cause mortality in men with prostate cancer treated by radical radiotherapy: 11-year follow-up data.

Abstract

e17003 Background: Low levels of serum testosterone are associated with higher all-cause mortality and cardiovascular mortality in some studies.(PMID: 18040028). But men with low levels of serum testosterone at reduced risk of developing prostate cancer. (PMID: 30077399). In middle-aged and elderly men, single point testosterone measurements have been shown to reliably reflect long term androgen exposure. [PMID: 1548361]. We assessed the interaction of pre-treatment serum testosterone levels and long term all-cause mortality in men with newly diagnosed prostate cancer. Methods: Prostate cancer pts enrolled in an ethics approved, prospective study evaluating a rating scale. Baseline serum testosterone levels were measured before the pts were commenced on androgen deprivation therapy and radical radiotherapy. Patient characteristics: Mean age 68.2yrs (median 69yrs); Age range category < 60yrs:14.9% , 61-70yrs:47.3% and > 71 yrs: 37.8%. Charlson Comorbidity Index (CCI) prevalence at baseline as follows CCI- Low risk : 39.9%; CCI-medium risk : 51.4%; CCI-high risk: 8.8%. PSA median level was 10.85 ug/L. Tumour was localised in a majority of pts- stage T1/T2 in 72% . Pts with M1 disease were not enrolled. Histology grade: Gleeson grade 6 in 7.4% and Gleeson grade 7 in 53.4% . Serum testosterone: Median 11.9 nmol/l; Mean 12.4 (Min-Max 2.6 to 26.7). Cox proportional hazards regression used to analyze the effect of baseline Serum Testosterone on long term overall survival, while controlling for covariates (age, cancer T stage, PSA category, Gleeson Category and Charlson Comorbidity index). Results: Lower levels of serum testosterone at baseline were associated with decreased all-cause mortality. There was no significant correlation between testosterone levels and age in our study sample of relatively elderly population (Pearson Correlation p = 0.362); There was no correlation testosterone levels and PSA levels(P = 0.821); no association between serum Testosterone levels and Gleeson Grade as well as T stage. (F = 0.78 ; P = 0.46 and F = 1.34 ; P = 0.26 respectively). There was also no significant variation in testosterone levels according to Comorbidity categorised by Charlson Comorbidity Index. (F = 1.31 P = 0 .26). Conclusions: Lower Serum testosterone levels were associated with decreased all-cause mortality in this prospectively collected survival data of non-metastatic Prostate cancer patients. [Table: see text]