Testosterone Therapy in Patients with Treated and Untreated Prostate Cancer: Impact on Oncologic Outcomes

Abstract

Purpose: Testosterone deficiency and prostate cancer have an increasing prevalence with age. However, because of the relationship between prostate cancer and androgen receptor activation, testosterone therapy among patients with known prostate cancer has been approached with caution.

Materials and methods: We identified a cohort of 82 hypogonadal men with prostate cancer who were treated with testosterone therapy. They included 50 men treated with radiation therapy, 22 treated with radical prostatectomy, 8 on active surveillance, 1 treated with cryotherapy and 1 who underwent high intensity focused ultrasound. We monitored prostate specific antigen, testosterone, hemoglobin, biochemical recurrence and prostate specific antigen velocity.

Results: Median patient age was 75.5 years and median followup was 41 months. We found an increase in testosterone (p <0.001) and prostate specific antigen (p = 0.001) in the entire cohort. Prostate specific antigen increased in patients on active surveillance. However, no patients were upgraded to higher Gleason score on subsequent biopsies and none have yet gone on to definitive treatment. We did not note any biochemical recurrence among patients treated with radical prostatectomy but 3 (6%) treated with radiation therapy experienced biochemical recurrence. It is unclear whether these cases were related to testosterone therapy or reflected the natural biology of the disease. We calculated mean prostate specific antigen velocity as 0.001, 0.12 and 1.1 μg/l per year in the radical prostatectomy, radiation therapy and active surveillance groups, respectively.

Conclusions: In the absence of randomized, placebo controlled trials our study supports the hypothesis that testosterone therapy may be oncologically safe in hypogonadal men after definitive treatment or in those on active surveillance for prostate cancer.

Keywords: neoplasm recurrence, local; prostate-specific antigen; prostatic neoplasms; risk assessment; testosterone.

Figure 1

Bar graph of serum testosterone levels before TT (dark grey) and after TT (light grey) for each treatment modality. Error bars represent 95% Confidence Intervals.

Figure 2

Line graph of median PSA changes over time while on TT, stratified by risk group.

Figure 3

PSA after TT initiation in the RP patient group. The N at each time point on the graph represents the number of patients with PSA data at said time point. Average time of last PSA drawn was 50 months. Mean was used in this case as median would result in an undetectable PSA at each time point.

Figure 4

PSA after TT initiation in the XRT patient group. The N at each time point on the graph represents the number of patients with PSA data at said time point. Average time of last PSA drawn was 39 months.

Figure 5

PSA after TT initiation in the AS patient group. The N at each time point on the graph represents the number of patients with PSA data at said time point. Average time of last PSA drawn was 32 months.