Prevalence of Hypogonadism in Low-Risk Prostate Cancer Survivors

Abstract

Increasing evidence, albeit small, suggests that testosterone replacement therapy can be cautiously considered in selected hypogonadal men treated with curative intent for low-risk prostate cancer and without evidence of active disease.

Hypogonadism is characterized by low testosterone levels that can result in symptoms such as reduced libido, erectile dysfunction (ED), fatigue, anemia, decreased bone density, decreased lean body mass, and increased body fat.^1,2 The Endocrine Society defines male hypogonadism as serum total testosterone level (T) < 300 ng/dL, a threshold at which the likelihood of most symptoms associated with hypogonadism increases.¹

Testosterone replacement therapy (TRT) is recommended for treatment of androgen deficiency in symptomatic men with unequivocally low serum T levels. Although current guidelines recommend against using TRT in men with a history of prostate cancer (PCa),^1,3 and the FDA has a black box warning against prescribing TRT for these men, little evidence suggests that TRT stimulates tumor growth in patients treated for low-risk PCa.^4–6 Retrospective studies and case series suggest that TRT can be safe in patients with low-risk PCa treated with radical prostatectomy, brachytherapy, or external beam radiation.^7,8 Despite observed increases in prostate specific antigen (PSA), TRT does not seem to increase PCa recurrence rates when used cautiously, even in men with high-risk disease.⁹

Almost 3 million men living in the U.S. have been diagnosed with PCa. While 1 in 7 men will be diagnosed with PCa during their lifetime, most men diagnosed with PCa do not die of it.¹⁰ More than 90% of patients with PCa have localized or low-grade disease that does not result in PCa-related mortality.¹¹ Active surveillance, brachytherapy, external beam radiation, and radical prostatectomy are considered appropriate monotherapy modalities for low-risk PCa.¹² With successful treatment of early PCa, primary care providers may increasingly encounter PCa survivors with or without symptomatic hypogonadism. Surprisingly, the prevalence of hypogonadism in men with low-risk PCa has not been reported.

The primary objective of this study was to estimate the prevalence of hypogonadism in low-risk PCa survivors who received curative treatment. A second objective was to examine the presence of hypogonadism among subgroups of patients. The authors hypothesized that the prevalence in this population may be high enough to support prospective trials designed to determine the safety of TRT in selected hypogonadal men with a history of PCa.

METHODS

This is a cross-sectional study conducted at the Edward Hines, Jr. VA Hospital (EHJVA) that included a convenience sample of 52 veterans aged 25 to 100 years who had been treated for low-risk PCa more than 12 months previously and were currently receiving medical care at EHJVA. Low-risk PCa was defined as tumors with a Gleason score ≤ 6 and PSA at diagnosis < 10 or a (clinical or pathologic) American Joint Committee on Cancer (AJCC) stage I or II. Patients were excluded if they had any of the following: Gleason score > 7, PSA at diagnosis > 10, prior TRT or androgen deprivation therapy, recurrent or active PCa, incomplete treatment of PCa, or history of breast cancer. The study was approved by the institutional review board at EHJVA.

Participant Identification and Recruitment

The EHJVA cancer registry provided a roster of about 600 patients who were diagnosed with AJCC stage I or II PCa after 2002 and had completed treatment by 2011. About 50% of the patients were excluded because they no longer were receiving care at EHJVA. More than 150 patients were excluded because they did not have upcoming appointments at EHJVA, they followed up at community-based outpatient clinics, or they did not meet the remaining parameters of the inclusion criteria.

Between April 2013 and August 2014, the authors approached 75 potentially eligible patients who had upcoming outpatient care appointments at EHJVA and invited them to participate in the study. After explaining the risks and benefits, 15 patients declined to participate. Although 60 eligible patients signed an informed consent, 6 later refused to consent to the medical record review. After the medical record review, 2 patients were excluded because their eligibility could not be confirmed (eg, missing tumor stage or PSA information). A total of 52 men were included in this study.

Data Collection

The study participants were asked to complete a brief validated Androgen Deficiency in the Aging Male (ADAM) questionnaire and provide a blood sample. The standard ADAM questionnaire consists of 10 yes/no questions concerning symptoms of androgen deficiency. A positive questionnaire was defined as a “yes” to any 3 questions.¹³ A blood sample was obtained between 7am and 9am for total and free testosterone level, follicle stimulating hormone, and luteinizing hormone (LH) to determine whether the hypogonadism was primary or secondary to a hypothalamic-pituitary process. Additional data were collected, including age, height, weight, race, presence of diabetes mellitus (DM), tumor characteristics at diagnosis (ie, PSA, Gleason score, AJCC stage), and first course of cancer treatment.

Main Outcomes and Measures

The primary outcome was hypogonadism. Patients with serum T < 250 ng/dL were identified as having hypogonadism (the lower limit of the reference range in the EHJVA laboratory assay 250–1,100 ng/dL). The type of hypogonadism was further categorized as primary (LH > 10.6 IU/L) or secondary (LH < 10.6 IU/L). Patients with a body mass index (BMI) > 30 kg/m² were considered obese.

The prevalence of hypogonadism was assessed overall and for subgroups of patients. The 95% confidence interval (CI) was calculated for each estimate. The correlation between BMI and serum T was also determined. Analyses were carried out using SAS 9.2 (SAS Institute Inc., Cary, North Carolina).

RESULTS

The mean age was 69.1 years and patients were primarily white (Table 1). Half the patients were obese, and 40% had DM. The majority of patients had been diagnosed with stage I PCa, had received radiation as their primary treatment, and had completed the PCa treatment more than 5 years ago. The mean ADAM score was 4.7 (a value > 3 indicates a positive screen for hypogonad symptoms). Based on laboratory results, 32.6% of patients had hypogonadism; 26.9% of the patients had secondary hypogonadism, and 6% had primary hypogonadism. Based on the ADAM questionnaire, the most common symptoms were erectile dysfunction, low libido, and low physical performance (Table 2).

Table 1.

Study Population Demographics^a

Parameter	No. (%)
Age (Mean 69 y, SD 7)
< 70 y	27 (51.9)
≥ 70 y	25 (48.1)
Raceb
White	32 (64.5)
Black	15 (28.8)
Other	4 (7.6)
Obese	26 (50.0)
Diabetic	21 (40.3)
AJCC stage
I	38 (73.0)
II	14 (26.9)
Cancer treatment type
Radiation only	36 (69.2)
Surgery only	15 (28.8)
Radiation and surgery	1 (1.9)
Time since cancer treatment
< 5 y	18 (34.6)
≥ 5 y	34 (65.3)
Hypogonadism
Primary	3 (5.7)
Secondary	14 (26.9)
None	35 (67.3)

Abbreviation: AAJCC, American Joint Committee on Cancer.

^an = 52.

^bRace was unknown for 1 patient.

Table 2.

Androgen Deficiency in the Aging Male Questionnaire^a

Symptoms	Positive Response,%
Decreased libido	73.1
Lack of energy	67.3
Decrease in strength and/or endurance	67.3
Lost height	23.1
Decreased “enjoyment of life”	32.7
Sad and/or grumpy	25.0
Erections are less strong	84.6
Recent deterioration in ability to play sports	36.5
Falling asleep after dinner	34.6
Recent deterioration in work performance	30.8

^an = 52.

The prevalence of hypogonadism did not differ by race, DM, or tumor stage (Table 3). The oldest age group had the highest prevalence, but it was not statistically significant. However, obese patients were significantly more likely to have hypogonadism than were nonobese patients (57.6% vs 7.6%, P < .001). The negative association between BMI and serum T was more apparent when the correlation (r = −0.37, P < .01) between these 2 factors were examined (Figure). Patients with a higher ADAM score (> 3) were more likely to have hypogonadism than patients with a lower score (40.6% vs 20.0%, P = .14; Table 3). The prevalence of hypogonadism was higher in patients who were treated with radiation compared with patients who had surgery (41.6% vs 13.3%, P = .05). However, obese patients were more likely to have been treated with radiation (Table 4). After stratifying by obesity status, there was no difference in hypogonadism by treatment type.

Table 3.

Hypogonadism by Select Characteristics

Characteristics	n	Any Hypogonadism, % (95% CI)	Primary, %	Secondary, %	None, %
Age, y
< 70	27	25.9 (9.4–42.4)	7.4	18.5	74.0
≥ 70	25	40.0 (20.1–59.2)	4.0	36.0	60.0
Race
White	32	37.5 (21.1–56.3)	3.1	34.3	62.5
Nonwhite	19	26.2 (9.1–51.2)	10.5	15.7	73.6
Obesea
Yes	26	57.6 (36.9–76.6)	11.5	46.1	42.3
No	26	7.6 (0.9–25.1)	0	7.6	93.3
Diabetes
Yes	21	33.3 (14.5–56.9)	4.7	28.5	66.6
No	31	32.2 (16.6–51.3)	6.4	25.8	67.7
AJCC stage
I	38	36.8 (21.5–52.1)	5.2	31.5	63.1
II	14	21.4 (0.0–42.9)	7.1	14.2	78.5
Treatment
Radiation only	36	41.6 (25.5–59.2)	8.3	33.3	58.3
Surgery only	15	13.3 (1.6–40.4)	0	13.3	86.6
ADAM score
≤ 3	20	20.0 (5.7–43.6)	0	20.0	80.0
> 3	32	40.6 (23.7–59.3)	9.3	31.2	59.3

^aP < .001.

Abbreviations: ADAM, Androgen Deficiency in the Aging Male; AJCC, American Joint Committee on Cancer; CI, confidence interval.

Figure.

Correlation Between Body Mass Index and Total Testosterone

Table 4.

Hypogonadism by Obesity and Cancer Treatment

Characteristics	n	Any Hypogonadism, % (95% CI)
Obese
Radiation only	21	61.9 (38.4–81.8)
Surgery only	4	50.0 (6.7–93.2)
All	26	57.6 (36.9–76.6)
Non-Obese
Radiation only	14	14 (1.7–42.8)
Surgery only	11	0
All	26	7.6 (0.9–25.1)

DISCUSSION

To the authors’ knowledge, this is the first study to assess the prevalence of hypogonadism in treated low-risk PCa survivors. In this sample of 52 men in an outpatient setting, about 1 in 3 had low testosterone levels. Three patients had primary hypogonadism, and 14 had secondary (hypogonadotrophic) hypogonadism. Obese patients were significantly more likely to have hypogonadism than were nonobese patients. There was no evidence of an association between PCa treatment and hypogonadism.

Estimates of prevalence of hypogonadism in the general population vary in the literature based on different patient demographics, study designs, and geography. A recent review found that the prevalence of hypogonadism in adult men ≥ 18 years had a range of 2.1% to 31.2% since the authors included studies that were population-based, community-based, and primary care or screening-based.¹⁴ The prevalence of hypogonadism was 15.0% to 78.8% in obese people and 21.4% to 33% in men with newly diagnosed PCa.¹⁴

Mulligan and colleagues estimated that 38.7% of men aged ≥ 45 years who visited primary care clinics in the U.S. had hypogonadism.¹⁵ They also found that the prevalence of hypogonadism, defined as serum T < 300 ng/dL, was significantly higher in men with obesity, DM, hypertension, hyperlipidemia, asthma or chronic obstructive pulmonary disease, and prostate disease. The prevalence of hypogonadism in this study was slightly lower (32.6%), but the authors used a different serum T level (< 250 ng/dL).¹⁵ Using the Mulligan and colleagues’ definition of serum T level, there was a prevalence of 40.3% in this study’s population of men with treated low-risk PCa.

Other patient characteristics were examined to determine whether the prevalence of hypogonadism differed by age, tumor stage, or mode of treatment. Testosterone levels drop in all men, 1% per year on average after age 30 years.¹⁵ Hypogonadism has been consistently found to increase with age and certain comorbidities. Zarotsky and colleagues found a 17% increase in the risk for hypogonadism with every 10-year increase in age.¹⁴ In this study, the authors did not find a significant difference in the prevalence of hypogonadism in this group by age or DM.

In the Prostate Cancer Outcomes Study (PCOS), the prevalence of erectile dysfunction was almost 90% in the prostatectomy and radiotherapy group after 15 years. It was unclear whether this was due to PCa and its treatment, the normal aging process, or a combination of factors, but no significant differences were observed between the treatment groups. However, it is unclear whether the PCOS men had hypogonadism, because testosterone levels were not reported.¹⁶

In the current study, the association between obesity and hypogonadism is consistent with epidemiologic evidence that suggests complex multidirectional interactions between testosterone, sex hormone-binding globulin, obesity, metabolic syndrome, and type 2 DM (T2DM), mediated by cytokines and adipokines. Several studies show that obesity adversely affects testicular function and is associated with a reduction in sex hormone-binding globulin, serum T, and free testosterone levels.^17,18 In addition, secondary hypogonadism has been shown to be higher in men who are overweight (25–29 kg/m² BMI) and obese (≥ 30 kg/m² BMI) compared with normal weight men.^19,20 Hypogonadism has also been shown to increase insulin resistance, thereby increasing the risk for developing metabolic syndrome, which is a precursor for cardiovascular disease (CVD) and T2DM.²¹ Furthermore, low testosterone concentration may be associated with increased insulin resistance, incidence of CVD events, anemia, and low bone density.^19,22,23

The relationship between obesity, metabolic syndrome, and androgen deficiency remains unclear because of the complex mechanisms involved in this association. Impairment of hypothalamic-pituitary function by decreased LH pulse amplitude, inhibitory effects of estrogen at the hypothalamus and pituitary, and the effects of leptin, ghrelin, and resistin, both centrally and on testicular Leydig cells, may explain lower testosterone level in obese males.^24,25 Recent studies have suggested a possible increased risk of CV events among groups of men prescribed TRT.^26,27 However, other studies demonstrate beneficial effects of TRT on CVD risk factors, and research over several decades suggests a strong beneficial relationship between normal T and CV health.²⁶ The evidence to suggest that TRT increases CV morbidity and mortality risks is poor,²⁸ but FDA is investigating the link between TRT and adverse CV outcomes.²⁹

Testosterone replacement therapy is recommended for symptomatic men with androgen deficiency to induce or maintain secondary sex characteristics and improve their sexual function, sense of well-being, muscle mass, strength, and bone mineral density. In a recent study supported by the National Institutes of Health (NIH) in men aged 65 years, increasing serum testosterone levels to mid-normal range for 1 year was associated with a moderate benefit in sexual function and improved mood, but there was no significant benefit in vitality (as measured by a fatigue scale) or walking distance.²⁸ In the NIH study, 4 men in the testosterone group and 1 in the placebo group received a diagnosis of PCa during or within the subsequent year of treatment, but the sample size was inadequate to reliably assess the effect of testosterone on the risk of PCa.²⁸

With increasing direct-to-consumer marketing of branded pharmaceutical products in the U.S., there is widening interest in testosterone levels and hypogonadism symptoms in middle-aged and older men. Indeed, 2.3 million patients received a prescription for testosterone in 2013, up from 1.3 million in 2010.^29,30 This change has become important because whereas TRT is standard therapy in symptomatic hypogonadal men, it has long been considered taboo for men with a history of PCa, regardless of disease status. Transdermal testosterone is contraindicated in men with carcinoma of the breast or known or suspected carcinoma of the prostate, according to the package insert.³¹ Androgens are contraindicated in men with known or suspected carcinoma of the prostate or breast, according to the testosterone cypionate injection package insert.³² Although TRT may increase serum PSA levels in some men, it often remains within clinically acceptable ranges and has not been shown to increase the risk for PCa. A recent observation from 3 registries of more than 1,000 hypogonadal men receiving TRT for up to 17 years concluded that TRT does not increase the risk for PCa.³³

Prostate cancer encompasses a heterogeneous collection of androgen-dependent and independent cells. Androgens have been known to play an important role in PCa biology, but this relationship is more complex than the traditional view that androgens stimulate PCa growth. More than 7 decades ago, Huggins and colleagues showed that disseminated PCa was inhibited by eliminating androgens by castration and activated by androgen injections.³⁴ Recently, the androgen hypothesis and the relationship of testosterone to PCa has been more clearly defined. Although it has been established that effective suppression of serum T levels with surgical or chemical castration remains an essential strategy in the management of advanced PCa, the assertion that testosterone causes growth of PCa has been challenged.^35,36

Recent studies have shown that there may be a more complex relationship between serum T and PCa risk than was previously established.³⁷ Although PCa cells have been shown to become androgen-independent as they progress into the castrate-resistant phase,⁹ several studies have indicated that low-serum T is associated with greater PCa risk and more worrisome features of PCa.^338,39 Hence, a saturation model has been proposed: Changes in serum T concentrations below the point of maximal androgen-androgen receptor (AR) binding will elicit substantial changes in PCa growth, as seen with castration or with serum T administration to castrated men.⁴⁰ However, once maximal androgen-AR binding is reached, the presence of additional androgen produces little further effect, suggesting that there is a limit to the ability of androgens to stimulate growth of PCa.⁴⁰

A meta-analysis of 45 articles studying the relationship between serum T and PCa risk has reported conflicting results.⁴ Eugonadal testosterone levels, whether physiologically or pharmacologically replaced, do not seem to promote PCa growth. It is unclear whether the timing of sex hormone exposure affects PCa or whether the cancer may influence blood levels of sex hormones. Since 2004, there have been case series totaling almost 150 men treated with prostatectomy, brachytherapy, or external beam radiation who have been safely treated with TRT.^6–9 These case studies suggest that after a thorough discussion of risks and benefits, TRT may be safer than previously thought for men who have been successfully treated for PCa, are deemed low risk for recurrence, and are monitored closely. However, no randomized controlled trials are available, and published guidelines recommend against starting TRT in patients with a history of breast or PCa.¹

Limitations

This study has limitations. The sample size is small and, therefore, it may not have had enough power to show differences in prevalence by subgroups. In addition, the population may not be representative of all male veterans who seek care in the VA or of men in other health care settings, because this was a convenience sample of VA patients who received most of their care in the VA. However, the prevalence of hypogonadism, in this outpatient population is very similar to that found among men seen in primary care clinics.¹⁴ Finally, because pretreatment serum T levels were lacking, the authors were unable to assess whether hypogonadism was present before surgery or radiation therapy, and unable to determine whether treatment had any effect on serum T levels.

The study has several strengths. First, to the authors’ knowledge, this is the first study to assess the prevalence of hypogonadism after patients with low-risk PCa have received treatment with curative intent. Second, the authors assessed hypogonadism in every patient by measuring serum T levels in the early morning when levels are known to be at their highest. Third, differentiating primary and secondary hypogonadism helped provide insight into the possible etiology of the low serum T levels. Fourth, this study was performed at a single institution that uses electronic medical records, with almost complete data on patient demographics, PCa treatment, and receipt of TRT or androgen deprivation therapy. Fifth, because this study had the participation of endocrinologists and urologists, the study design helped answer questions pertinent to both medical and surgical specialties.

CONCLUSION

As life expectancy increases, many survivors of treated PCa present with symptoms of hypogonadism associated with low serum T levels and request TRT. The prevalence of hypogonadism before or after treatment for PCa in this population is not known. This study suggests that many low-risk PCa survivors have hypogonadism. Because hypogonadism negatively impacts quality of life by increasing the risk for sexual dysfunction, mood disturbances, bone fracture, development of metabolic syndrome, frailty, and decline in the feeling of general well-being and may have significant deleterious effects on other body systems, consideration for treatment is warranted.

Patients with PCa may be untreated because the safety of TRT in this population is unknown. Clinical practice guidelines caution against using TRT in this population, and recent literature questions the benefits and risks associated with the long-term safety of TRT, particularly in older men.^28,41 Although further studies are necessary before definitive conclusions can be drawn, increasing evidence, albeit small, suggests that TRT can be cautiously considered in selected hypogonadal men treated with curative intent for PCa and without evidence of active disease. However, because obese patients are at higher risk for aggressive PCa and mortality, it is unclear whether obese PCa survivors have an additional risk in regard to TRT.^42,43 To help clinicians provide information and care for their patients, appropriately designed prospective randomized studies using a collaborative approach and long-term follow-up are urgently needed to determine the safety of TRT in hypogonadal men with a history of low-risk PCa.