by Stephanie T. Page, MD, PhD. 

In the United States in 2009, prostate cancer will be diagnosed in over 190,000 men and will be responsible for over 27,000 deaths (58).  One of the Initial studies that examined the relationship between testosterone and the prostate was performed by Huggins and Hodges in 1941.  This study found that removal of testosterone reduces the growth of prostate cancer cells (59).  In addition, studies in hypogonadal men treated with TRT have demonstrated increases in prostate size and prostate-specific antigen (PSA) values relative to pretreatment or placebo-treated hypogonadal men (60).  It is therefore a common belief that the use of testosterone supplementation, especially in older men at risk for prostate disease may have prostate related side effects.  Much of the data examining the relationship between testosterone and prostate health, however, conflicts with this belief. 

A more detailed examination of the response of the prostate to TRT was carried out by Behre et al, who found that administration of testosterone resulted in increase in both PSA and prostate volume in hypogonadal men that matched, but did not exceed, was was seen in healthy, eugonadal, age-matched controls (51).  Regarding the relationship between endogenous testosterone levels and prostate cancer risk, a meta-analysis of 18 prospective studies that examined the risk of prostate cancer as a function of endogenous testosterone concentration found that elevated endogenous testosterone levels were not associated with an increased risk of prostate cancer (62).  

Other, more limited data, show that the frequency of prostate cancer may increase as total testosterone levels decrease (63).  The effect of testosterone supplementation on prostate health has also been examined.  A randomized, double-blind study of 60 healthy men between the ages of 60 and 75 who were made hypogonadal with a GnRH agonist examined the effect of varying doses of testosterone supplementation (25, 50, 125, 300, 600 mg/week) on multiple physiological measures (64).  Testosterone administration resulted in a dose-dependent increase in serum free testosterone levels but, more importantly, did not result in a dose-dependent increase in PSA levels (64). 

The direct effect of androgens on the prostate has been examined in studies of 5 cx-reductase nhibitors, which function by blocking the conversion of testosterone to the more potent androgen dihydrotestosterone (65).  In the prostate, dihydrotestosterone is found at a higher concentration than testosterone, whereas in the plasma, the converse is true (66).  The Prostate Cancer Prevention Trial (PCPT), which consisted of 18,882 men who were at least 55 and had a PSA level less than 3.0 ng/mL, examined the effect of the 5 cx-reductase inhibitor finasteride on the development of prostate cancer over a period of 7 years (65).  The prevalence of prostate cancer was reduced in patients who received finasteride (P>0.001), but finasteride administration was also associated with a significant increase in high-grade (Gleason grade 7 or higher) tumors compared with placebo (P>0.001) (65).  The PCPT authors explained that the increase in high-grade tumors was due to apparent finasteride-induced histologic changes that resulted in improved biopsy sensitivity for high-grade tumors (67). 

The effects of another 5 cx-reductase inhibitor, dutasteride, on prostate health was examined in the Reduction by Dutasteride of Prostate Cancer Events (REDUCE) rrial.  The 4-years, double-blind REDUCE trial consisted of 8,220 men between the ages of 50 and 75 with PSA scores between 2.5 and 10ng/mL and a single negative biopsy 6 months prior to enrollment (68).  The incidence of prostate cancer in patients who were randomized to receive dutasteride was 13.4% after 4 years of treatment (68).  The incidence of prostate cancer in the placebo group was 17.2% after 2 years of treatment and an additional 11.8% after 4 years of treatment (68).  The number of high-grade tumors did not differ between the treatment groups (68).  These studies suggest that a significant altercation in dihydrotestosterone levels within the prostate can have a significant effect on prostate cancer risk.  It is therefore important to understand how changes in plasma androgen levels are a result of TRT translate into changes in androgen levels in the prostate. 

Recent studies have shown that a significant reduction in plasma androgen levels does not result in the complete abolition of androgen levels in the prostate.  One of the first trials to support such a hypothesis compared prostate cancer biopsies from 22 men with recurrent prostate cancer who had received androgen deprivation therapy for an average of 37 months with benign prostate biopsies from an additional 48 men (69).  Levels of testosterone in the prostate tissue did not differ between groups, and dihydrotestosterone levels - though reduced in those who received androgen deprivation therapy (P=0.000007) - were found at a concentration of 1.45 nM, whereas control prostate tissue had a dihydrotestosterone level of 8.13 nm (69). 

In another study, it was found that in healthy men (N=13; 35-55 years) who were medically castrated (through the use of GnRH antagonist) for 28 days, both testosterone and dihydrotestosterone levels were significantly reduced in the serum (P<0.05 vs baseline) and prostate tissue (P<0.05 vs baseline) (70).  Despite the significant reduction in dihydrotestosterone in prostate tissue, a high concentration of dihydrotestosterone (2ng/mL, a level that is higher than normal serum levels) was still found in the prostate after medical castration, and the decrease in testosterone in the prostate was not as large as what was seen in plasma (70).  Analysis of PSA, prostate epithelial proliferation, apoptosis, and androgen receptor expression did not differ between those who were medically castrated and the placebo group, suggesting that residual androgens within the prostate were able to support androgen-regulated biologic activity (70). 

In men with hypogonadism, the effect of testosterone supplementation on androgen levels in the prostate has been investigated.  In a study of 44 men with serum testosterone levels less than 300 ng/dL, testosterone supplementation significantly increased serum testosterone and dihydrotestosterone levels (P<0.001) to within the normal range for young men, but did not increase prostate tissue levels of testosterone or dihydrotestosterone compared with baseline values (55).  There were no treatment-related changes in prostate tissue histology, tissue biomarker expression, or gene expression, nor was there a change in the incidence or severity of prostate cancer (55). 

The minimal effect of modifying plasma androgen levels on physiological and biochemical changes in the prostate indicates that there may be a buffering system in the prostate.  The purpose of a buffer system within the prostate is to allow for maintenance of intraprostatic hormone levels, despite changes in serum androgen concentration.  An alternative, but not mutually exclusive, mode of prostate responsiveness to androgen manipulation has recently been proposed by Morgentaler and Traish (71,72).  This "saturation model" suggests that prostate size and androgen-regulated cellular processes increase directly with serum testosterone concentration until serum testosterone concentrations reach low-normal levels (71,72).  Once eugonadal serum testosterone concentrations are achieved, prostate responsiveness to androgens begins to plateau with further increases in serum testosterone concentration, since the androgen receptors within the prostate are already saturated (71, 72).  If correct, these models help clarify what may be happening to the prostate of otherwise healthy hypogonadal men who are receiving TRT; restoration of eugonadal testosterone levels during TRT does not increase prostatic growth and disease because of the saturation effect. 

In hypogonadal men with a history of prostate cancer, there are little data on the effect of TRT on prostate health of the risk of prostate cancer recurrence.  In a total of 116 patients across 7 different studies that looked at the effect of testosterone use in hypogonadal men with a history of prostate cancer, it was determine that 1 patient had a recurrence of prostate cancer after TRT (71).  However, these data are derived from a case series, as data from randomized trials are lacking.  WHen considering TRT for hypogonadal patients, The Endocrine Soceity recommends that a urological examination to be perform in patients with a PSA greater than 3 ng/mL prior to initiation treatment (5).  In patients who are beginning TRT, a digital rectal exam and PSA should be repeated 3 months after the start of therapy and then annually or at a frequency in accordance with evidence-based guidelines for prostate cancer screening (8).  

Summary 

• The relationship between serum androgen concentrations and prostate health in men has not been conclusively determined. 
• 5 cx-reductase inhibitor studies suggest that lowering levels of dihyrotestosterone, but not testosterone, reduces the risk of developing prostate cancer in older men. 
• Physiological and biochemical data suggest that the increase in serum testosterone that results from TRT have limited effects on the prostate.  The administration of TRT does not appear to alter testosterone or dihydriotestosterone levels with the prostate nor does it cause any demonstrable pathological changes to prostate tissue after 6 months of treatment. 
• It is currently uknown how hypogonadism should be treated in patients with a history of prostate cancer; therefore, such patients should be enlisted in a clinical trial when possible and counseled extensively regarding the possible risks of exacerbating residual prostate disease prior to initiating therapy.