Relationships of Testosterone and Prostate Cancer: A 2011 Perspective
It is estimated that more than one third of men over age 45 suffer from hypogonadism (defined as a total testosterone level below 300 ng/ml). Perhaps even more disturbing to health care providers invested in men’s health is the fact that only 4%-5% of these men are receiving treatment for the condition.1 While hypogonadism may have a negative effect on quality of life and overall health, the unspoken question of how testosterone replacement may affect men who have been or will be diagnosed with prostate cancer (PCa) looms in the background. The objective of this review is to better define the role that testosterone plays in the diagnosis, prognosis, and growth of PCa and to review the current status of treatment for hypogonadism in the PCa patient.?
Background?
Testosterone is a critical hormone in the regulation of metabolic, sexual, and developmental functions. It not only facilitates the development of secondary sex characteristics such as deeper voice and male hair growth patterns, but also fuels bone growth and lean muscle mass development. It has been postulated to play a role in cognition as well, and research has begun to associate testosterone with brain development, concentration, and possibly memory.2 Symptomatic hypogonadism can dramatically decrease the quality of a man’s life by decreasing libido, affecting erectile function, precipitating hot flashes and gynecomastia, as well as causing fatigue and depression and other metabolic and cognitive derangements.3,4,5,6 ?
Despite the health benefits men reap from testosterone, there continues to be understandable concern regarding the safety of testosterone replacement therapy (TRT) given the role the hormone plays in advanced PCa. The sordid past of testosterone began in 1941 when Charles Huggins demonstrated that metastatic PCa was activated by androgens and inhibited by their absence.7 In spite of the advances medicine has made since that time, it is still androgen deprivation that continues to provide men with advanced PCa the relief they seek from pain and other associated morbidities by promoting regression of metastatic disease. ?
Health care providers must question whether TRT does more harm than good for men who have definitively treated PCa who also suffer from hypogonadism. New theories regarding the biology of testosterone are continually evolving and informing the discussion. Morgentaler and Traish have introduced the hypothesis of the "saturation model" to explain seemingly disparate roles of testosterone in men’s overall health with respect to PCa. They suggest that testosterone relies on an androgen receptor (AR) to exert biologic effects on PCa, and because of a finite number of ARs, it is possible to saturate the receptors. Any increase in testosterone above this level of saturation will not result in greater biologic effects. This model explains the exquisite sensitivity of advanced PCa to even small amounts of testosterone and provides a rationale for why castrate levels of testosterone are effective for androgen deprivation while increasing levels of testosterone above castrate levels seem to have a negligible effect on PCa growth.8 This theory also suggests that TRT in hypogonadal men will not exponentially fuel a smoldering cancer. ?
Given the surge of men diagnosed with PCa in the post-PSA era combined with the increasing incidence of men diagnosed with hypogonadism, the debate regarding the role of TRT is more salient than ever. Current guidelines clearly state that androgens are contraindicated in patients with known PCa.9 The evidence to support this position, however, is controversial.?
TRT and Prostate Cancer Risk?
Rhoden et al sought to determine whether TRT could precipitate PCa development in men with high-grade prostatic intraepithelial neoplasia (HGPIN) of the prostate. The authors identified 75 hypogonadal men who had prostate biopsies prior to initiating TRT for 12 months. Fifty-five men had negative prostate biopsies (HGPIN-), and 20 had HGPIN identified on biopsy (HGPIN+). There was no statistical difference in PSA values of HGPIN- or HGPIN+ men both before and after one year of TRT. Six men underwent rebiopsy for a PSA increase of more than 1 ng/ml, and no biopsy identified PCa. One man (HGPIN+) developed an abnormal DRE during the course of the study and a rebiopsy led to a diagnosis of Gleason 7 cancer.10 Given that 15%-25% of men have been found to have prostate cancer within three years of identifying HGPIN on biopsy, the finding of one incidence of PCa in 20 HGPIN+ men over one year suggests TRT does not increase the development of PCa in this subgroup.11?
Coward et al provide additional valuable insight into the role of TRT and PCa risk. Eighty-one hypogonadal men were followed for a mean of 34 months while receiving TRT. Four men (4.9%) were diagnosed with PCa at a mean of 33 months after initiating TRT, an incidence no greater than found in the general population. Notably, PSA did not significantly increase in the 95.1% of men who did not develop PCa, while the PSA did increase in the men diagnosed with PCa. Although a small study, the results strongly suggest not only that TRT does not increase the risk of developing PCa, but also that TRT does not eliminate the utility of PSA values in identifying PCa when it does occur.12 ?
Likewise, Shabsigh et al demonstrated in their meta-analysis of 44 publications that there is no correlation between TRT and increasing PCa risk or increasing the risk of higher Gleason grade disease.13 Marks et al provide some interesting insight to understanding why TRT might not affect PCa risk. In their double-blind, placebo-controlled trial evaluating treatment of hypogonadal men with TRT, they measured both serum and prostate tissue levels of testosterone and dihydrotestosterone (DHT). Despite increases in serum levels of both, there was no increase in intraprostatic testosterone or DHT levels.14?
While these studies are reassuring, we cannot draw firm conclusions given the need for additional research in this area. Clearly, the restricted duration of follow-up for the development of PCa after initiating TRT is just one resounding limitation in all studies to date. These studies, however, do serve to underscore the lack of evidence-based medicine physicians face when attempting to appropriately council patients regarding the risks and benefits of TRT. ?
Testosterone and Risk of More Aggressive Prostate Cancer?
At present, the relationship between testosterone level and risk of PCa is unclear. Some papers suggest that low testosterone levels predict more aggressive disease and poorer prognosis in men with advanced PCa. Hoffman et al demonstrated in 117 PCa patients a trend between low testosterone and more extensive disease. There was a statistically significant increase in Gleason 8 disease among men with low testosterone.15 Interestingly, Chodak et al demonstrated that the higher the serum testosterone prior to treating men with metastatic PCa with a gonadotropin-releasing hormone analogue, the better the survival.16 Massengil et al investigated the relationship of pre-operative testosterone and PCa prognosis through a retrospective analysis of 879 men treated with radical prostatectomy (RP). They demonstrated that men with pT3-T4 disease had lower pretreatment total testosterone levels than those with organ confined cancer (pT1-T2), but also noted that total testosterone was not a significant predictor of biochemical recurrence.17 Isom-Batz et al retrospectively reviewed the pre-treatment testosterone level of 326 men who underwent RP and did not receive adjuvant therapy. They identified a statistically significant association between lower testosterone levels and higher clinical stage and biopsy grade. This study, too, found no association between testosterone levels and the 41 men in their study who experienced biochemical recurrence.18 Schatzl et al also independently demonstrated a correlation between patients with lower testosterone and the diagnosis of higher Gleason grades in 156 PCa patients. Interestingly, gonadotropins were also lower in men with low testosterone in this study, suggesting that PCa may suppress the hypothalamic-pituitary-gonadal axis or that this is not a primary leydig cell-related issue.19 These studies underscore the possibility that low pretreatment testosterone may portend a poor prognosis and low testosterone does not appear to be protective against the development of biochemical recurrence.?
The pathophysiology explaining these findings remains to be elucidated. Some theorize that PCa cells can inhibit testosterone production, and this hypothesis would suggest that low testosterone levels could play a role as a marker for high grade prostate cancer.15 PSA expression is androgen dependent, and low testosterone may mask an otherwise elevated PSA in the context of PCa; this in turn may cause a delay in diagnosis that would allow the cancer to grow to a more advanced stage. ?
Alternatively, perhaps testosterone exerts growth maturation effects on PCa cells and can promote differentiation into a less aggressive cancer, promote senescence, or even stimulate apoptosis. These ideas allow for the possibility that testosterone itself could actually be therapeutic if its administration could attenuate PCa growth. Some researchers have touted 5α-reductase inhibitors (5ARIs), which block the conversion of testosterone to the more potent androgen receptor binder dihydrotestosterone (DHT), for PCa chemoprevention.20,21 Interestingly, 5ARIs, because of their mechanism of action, have been shown to increase the intraprostatic level of testosterone up to 20 times the level found before instituting 5ARI therapy.22,23 Additional research addressing whether it is not only the decrease in DHT, but also the elevation in intraprostatic testosterone, which plays a potential tumor suppressive role is certainly needed. Notably, both PCa cell lines and a PCa mouse model have been shown to exhibit androgen suppression of cell and tumor growth, suggesting a cellular mechanism for growth inhibition that remains to be identified and exploited as a novel PCa treatment method.24,25 ?
Whether testosterone can repress growth of metastatic PCa has already begun to be investigated in nascent clinical studies. In a pioneering Phase 1 clinical trial, Morris et al tested the safety of high-dose exogenous testosterone in 12 men with castration-resistant metastatic PCa. None of the patients in the study had pain flares or grade 3 or 4 toxicity associated with treatment. One man with known epidural disease did develop cord compression without neurologic impairment that was treated with radiation.26 While this study was not designed to ascertain efficacy, the safety profile it demonstrated has opened the door for studies which can examine clinical response. ?
Conversely, several published studies suggest that higher serum testosterone level increases PCa risk, correlates with more aggressive PCa, or even has no relationship with PCa risk. Gann et al found, through a prospective case-controlled study nested within the Physicians’ Health Study, an increased risk of PCa in men with higher (albeit within normal range) serum testosterone levels.27 Pierorazio et al found a statistically significant increase in calculated free testosterone and calculated free testosterone index in 26 men aged 65 and older with high-risk PCa when compared with 355 age-matched men without PCa or with low-risk PCa.28 Koo et al identified 120 men with PSA levels above 10 ng/mL who underwent prostate biopsy; 35 men were diagnosed with PCa. This study found no correlation between testosterone level (whether hypogonadal or eugonadal) and PCa risk, PSA elevation, or PCa aggressiveness.29 ?
The sheer variety of conflicting data that often exonerates, sometimes condemns, and occasionally discounts a role for testosterone with regard to PCa risk and growth only serves to clarify the need for more data before a definitive conclusion can be drawn.?
TRT in Patients with Known Prostate Cancer?
A growing body of evidence suggests a margin safety for TRT in hypogonadal men with treated PCa. Several studies have demonstrated symptomatic improvement in hypogonadism without the development of biochemical recurrence in PCa patients after RP. In 2004, Kauffman and Graydon presented the first published report of seven symptomatic hypogonadal men who had undergone RP and were treated with TRT. All men were eugonadal preoperatively and had Gleason 6 or 7 localized disease. In follow-up that ranged from 1-12 years, PSA levels remained undetectable and there was no evidence of clinical recurrence. Notably, one patient had positive surgical margins at the apex and his PSA remained undetectable after 12 years of TRT.30 ?
Agarwal et al also demonstrated a similar safety profile in their reports of 10 men who underwent RP and received TRT. After a median follow-up of 19 months, all had symptomatic benefit from TRT without any evidence of biochemical recurrence.31 Khera et al evaluated 57 men who received TRT after RP. After treatment periods that ranged from 1-136 months, none of the men had an increase in their PSA values.32 Sathyamoorthi et al presented data from 133 patients, including some patients from the Khera study, followed for a mean of 12 months. There were no PSA recurrences and no statistically significant increases in PSA. Sixteen of these patients had high-grade disease (Gleason 8) or positive margins.33 ?
The results of studies in hypogonadal men with PCa treated with brachytherapy and/or external beam radiation are equally heartening, and also reveal a margin of safety for TRT in men who have close follow-up. Sarosdy followed 31 men with regular PSA and testosterone levels for a median of five years after brachytherapy with or without external beam radiation treatment for Gleason 5-9 PCa who were receiving TRT for symptomatic hypogonadism. Only one man had an increase in his PSA values after starting TRT. Although he continued his TRT, his PSA value ultimately declined. The most recent serum PSA levels were below 0.1 in 23 patients (74%), 0.1-0.5 in seven (23%), and 0.5-1.0 in one (3%). No patients, however, underwent prostate biopsy to exclude cancer recurrence.34 Morales et al followed five men with PCa who underwent EBRT and received TRT for a median of 14.5 months. One man had a transient increase of his PSA above 1.0, but no patient had a PSA above 1.5 while receiving TRT.35?
Recently, Morgentaler et al presented their data with testosterone replacement in patients on an active surveillance protocol for PCa. Fourteen patients were followed for a mean of 23 months. Thirteen patients had prostate biopsies showing a maximum Gleason score of 6 and one patient had evidence of Gleason score of 7. The mean PSA in the group increased from 3.7 to 5.5 over 23 months of follow-up. These short term data suggest that the rate of disease progression is not higher than the natural history of the disease. One patient without evidence of PSA increase desired a repeat biopsy, which revealed one core of Gleason 7. The final pathology after RP was Gleason 6 cancer involving 5% of the gland. 36 ?
The limitations of these studies are clear, and include small sample size, non-randomized design, and relatively short follow-up. The aim of these studies is to begin the process of exploring the possibility of establishing the safety of TRT in men who have been diagnosed with PCa. ?
Current Guidelines Regarding TRT?
The guidelines for treatment published in 2008 by the writing group for four international andrology societies concluded that: ?
"Men successfully treated for prostate cancer and suffering from confirmed symptomatic hypogonadism are potential candidates for testosterone substitution after a prudent interval if there is no clinical or laboratory evidence of residual cancer. The risks and benefits must be clearly discussed with and understood by the patient and follow up must be particularly careful."9 ?
The 2010 Endocrine Society Guidelines remain clear that "testosterone is contraindicated in prostate cancer." However, in the body of they report, they mention the following:?
"Although some clinicians have suggested that patients with organ-confined prostate cancer who have undergone radical prostatectomy and have been disease-free 2 or more years after radial prostatectomy and who have undetectable PSA levels may be considered for testosterone replacement on an individualized basis, the lack of data from randomized trials precludes a general recommendation."37 ?
This shift in discussion stands in stark contrast to the historical perception that any history of prostate cancer is an absolute contraindication to TRT.?
Monitoring TRT?
Prior to instituting TRT and during its utilization, prostate evaluation is critical for men age 40 and older. Baseline evaluation should include a DRE, measurement of the serum PSA, and assessment of bladder outlet obstruction (BOO) with either the American Urological Association Symptom Score or the International Prostate Symptom Score questionnaire. The DRE, serum PSA, and BOO evaluation should be repeated at 3, 6, and 12 months and then every 6-12 months.38 PSA increases have been observed with TRT and have been reported in the range of 0.2-0.5, observed over 3-12 months, whereas other studies have demonstrated no such increase.39,40,41 PSA rises have been correlated with benign prostatic hyperplasia, however, and not with the development of overt cancer. Notably, multiple studies have not demonstrated worsening of voiding dysfunction in men on TRT, and complications such as urinary retention have not occurred at higher rates than in men receiving placebo.38 The subsequent rise in PSA and prostate volume have, in fact, been demonstrated to rise to the level of age-matched controls.41 ?
Serum testosterone levels should clearly be followed to evaluate the effectiveness of the TRT. The initial serum testosterone level should be checked 2-4 weeks after initiating therapy. The frequency of evaluating subsequent serum testosterone levels will be dependent on the mode of delivery of testosterone. Because one of the instigating factors for prescribing TRT is to manage the osteopenia and osteoporosis that develops in hypogonadal men, bone mineral density and a bone scan of the lumbar spine or femoral neck should be performed every 1-2 years. Elevated hematocrit can become problematic for men on TRT and should be checked at baseline, after three months of TRT, and then yearly. If the hematocrit is above 52%, the dose of TRT should be reduced or discontinued altogether.38?
Conclusions?
In the 70 years since its identification as a PCa growth promoter, testosterone has become inextricably associated with this disease. The paradigm of limiting the use of testosterone in patients at risk for or after treatment of PCa is changing. The authors believe that with proper informed consent, testosterone may be offered to patients treated for PCa with the understanding that only small, non-randomized studies with short follow-up have shown no worse outcomes. Long-term, randomized trials will help further elucidate the safety and efficacy of this therapy. n?
Kelly Chiles, MD is a resident in the Division of Urology at the University of Connecticut Health Center in Farmington, where Stanton Honig, MD, is Associate Clinical Professor of Surgery. Dr. Honig also practices at the Urology Center in New Haven, Conn. and is a member of the Renal & Urology News editorial advisory board.
References?
1. Mulligan T, Frick MF, Zuraw QC, et al. Prevalence of hypogonadism in males aged at least 45 years: the HIM study. Int J Clin Pract. 2006;60:762-769.?
