Testosterone Therapy in Hypogonadal Men with Prostate Cancer

Prostate Cancer as a Contraindication for Testosterone Therapy

Numerous cross-sectional and longitudinal studies demonstrate that serum testosterone levels decrease with advancing age while the incidence of symptomatic male hypogonadism increases and may be as high as 500,000 new cases annually in men aged 40–69 years in the United States. Similar data have been shown for men in Australia and Europe. The implications and consequences of hypogonadism are increasingly diagnosed; they are wide-ranging and include decreased quality of life, energy and libido, loss of muscle mass, cognitive impairment, sexual and erectile dysfunction, poor attention span, depressed mood, and decreased cardiovascular and bone health. Treatment with testosterone replacement therapy can alleviate these signs and symptoms.

Young men (<50 years of age) comprise approximately 2% of new prostate diagnoses in the United States. The median age at prostate cancer diagnosis is 68 years of age, and 63% of patients are diagnosed at 66 years or older. It is reasonable to conclude that there are a large percentage of hypogonadal men who have also undergone treatment for prostate cancer. In addition, patients undergoing treatment for prostate cancer are at increased risk of depression, erectile and sexual dysfunction, and decreased libido. Men with low serum testosterone levels are thus at high risk of experiencing these symptoms, and may experience significant benefit from testosterone replacement therapy. Unfortunately, these are exactly the men in whom testosterone replacement therapy has been traditionally withheld.

The historical basis for withholding testosterone replacement therapy in patients with prostate cancer, even after definitive treatment for local disease, stems from the thinking that the prostate is an androgen-dependent organ and prostate cancer uses testosterone as a substrate or “fuel” for growth. Indeed, many currently practicing urologists were taught during training that administering testosterone to patients with prostate cancer is akin to “throwing gasoline on a fire” or “feeding a hungry tumor.” The original, oft-cited assertion that testosterone activates prostate cancer derives from Huggins’ seminal studies more than 65 years ago, in which he showed that castration resulted in a rapid reduction in serum acid phosphatase and testosterone treatment caused a rapid increase in acid phosphatase. These were conducted prior to discovery of the androgen receptor, accurate serum testosterone assays, and prostate-specific antigen (PSA). Subsequently, in 1967, Prout and Brewer reported that 5 of 10 men with recurrent disease after castration experienced progression or death when given several weeks of testosterone supplementation. A 1981 study by Fowler and Whitmore demonstrated “unfavorable” outcomes in 45 of 52 men with metastatic disease administered exogenous testosterone. The result of these trials was near-universal acceptance of the view that increased serum testosterone leads to prostate cancer growth and progression; accordingly, prostate cancer has long been considered a contraindication to testosterone supplementation, even after primary treatment for organ-confined disease.

New Thinking: the Saturation Model

During the past decade, a number of authors began to question the validity of the link between serum testosterone level and prostate cancer activity, noting that the link between Huggins’ castrated men and current hypogonadal men treated for prostate cancer is tenuous, with no clear evidence of increased risk of recurrence for men successfully treated for primary prostate cancer. Given the considerable overlap of patients with hypogonadism and prior treatment for prostate cancer, as well as increasing recognition of the efficacy and benefits of testosterone therapy, a limited yet significant body of literature has been published focusing on the safety and role for testosterone supplementation in men who have undergone treatment for prostate cancer.

In addition to questioning whether it is justifiable to withhold testosterone therapy in men with prostate cancer status post definitive treatment, Morgentaler and Traish proposed a revised view of the effect of testosterone on prostate cancer stimulation and growth: the saturation model. Noting that androgens have a limited capacity to stimulate prostate cancer progression, they described a saturation point such that levels of testosterone at or below the near-castrate range are associated with a steep T-dependent rate of growth, while minimal or no growth is seen above the level of saturation. Morgentaler and Traish argue that administration of exogenous testosterone to hypogonadal (but not castrate-level) men would constitute serum fluctuation of testosterone at levels above the saturation point, which should not impact the rate of cancer recurrence or growth, yet may provide marked benefit to hypogonadal men who have undergone treatment for their prostate cancer.

In a rat model, Wright et al. demonstrated that intraprostatic androgen levels and prostate mass are sensitive to serum testosterone at near-castrate levels, but sensitivity decreases as levels increase. A small, randomized controlled trial conducted by Marks et al. on 44 men with late-onset hypogonadism assessed the effect of androgen replacement therapy on prostate tissue. Forty men underwent both randomization and follow-up biopsy at 6 months. Testosterone replacement increased serum levels but had no effect on intraprostatic androgen and biomarker levels, gene expression, or prostate cancer incidence ( Figure 67.1 ). Other previously published studies of androgen binding capacity in animal and human prostate tissues have supported the possibility of a saturation model, demonstrating maximal binding (saturation) of androgen to androgen receptors in the 3nM or less range. Viewed in consideration of these findings, the Huggins, Prout, and Fowler cohorts’ unfavorable responses very well may have been experienced by patients who had undergone castration prior to testosterone treatment – and were thus functioning at levels of serum testosterone lower than the saturation point with unbound androgen receptors available to be activated by exogenous testosterone.

Muller et al. provided further compelling evidence in support of the saturation model in 2012 when they reported on the relationship of serum testosterone and dihydrotestosterone to prostate biopsy results in the reduction by dutasteride of prostate cancer events (REDUCE) trial. Trial entry requirements included PSA between 2.5 ng/mL and 10.0 ng/mL and prior negative biopsy. The authors analyzed the 3255 men randomized to the placebo arm who underwent at least one of the planned biopsies at 2 and 4 years with regard to baseline serum testosterone and dihydrotestosterone values. No significant association between prostate cancer and serum androgens was identified. Cancer rates were similar between men with normal testosterone and men with hypogonadism defined as <10 nmol/L or 288 ng/dL (25.5% vs. 25.1%, respectively; p = 0.831). The authors included a LOWESS plot in their paper, presented later, stating “prostate cancer detection between men with low compared with normal baseline testosterone was similar, perhaps because the optimal threshold above which testosterone becomes saturated may be ≤10 nmol/L.” The rates of prostate cancer are seen to plateau at a value below the normal range.

In an editorial in the same journal issue as the Muller study, Morgentaler wrote: “Muller et al. provide the final nail in the coffin for what had been a guiding principle of uro-oncology for >70 years: the androgen hypothesis.” Morgentaler went on to emphasize the importance of this study, as “the failure to find increased (prostate cancer) rates associated with higher serum androgens based on biopsies in a large at-risk population removes the last possible hope to those who wish to hold on to a disproved theoretical notion from a premodern era.” The title of the editorial succinctly summed up the shift in thinking that has taken place regarding the use of testosterone therapy in patients with a diagnosis prostate cancer over the past decade: “Goodbye androgen hypothesis, hello saturation model.”