Guidelines for Testosterone Therapy for Men: How to Avoid a Mad (T)ea Party by Getting Personal

“Take some more tea,” the March Hare said to Alice, very earnestly.

“I've had nothing yet,” Alice replied in an offended tone, “so I can't take more.”

“You mean you can't take less,” said the Hatter. “It's very easy to take more than nothing.”

“Nobody asked your opinion,” said Alice.

“Who's making personal comments now?” the Hatter asked triumphantly.

—A Mad Tea-Party in Alice in Wonderland (1)

In this issue of JCEM, the original authors of The Endocrine Society's inaugural guideline on “Testosterone Therapy in Men with Androgen Deficiency Syndromes,” published in 2006 (2,3), offer an update. What is the value of updating these guidelines just 4 yr later?

As the authors noted in the original 2006 guidelines, there were no randomized, placebo-controlled trials of the effects of testosterone therapy on clinically important outcomes such as mortality, quality of life, cardiovascular events, fragility fractures, or cognitive function in men. There was also a paucity of data about the effects of testosterone therapy in older men, men taking corticosteroids, and other clinical scenarios where men commonly have low serum testosterone levels and symptoms or signs that might be due to androgen deficiency. Unfortunately, there are still no published, long-term, large placebo-controlled trials examining clinically meaningful outcomes in hypogonadal men. As with the original 2006 guidelines, the evidence for the 2010 guidelines is weak. On a rating scale of very low, low, moderate, and high quality, all of the recommendations were based on evidence judged to be very low or low, and the majority of the recommendations were based on very low-quality evidence.

This deplorable state of affairs occurs nearly a decade after the publication of the Women's Health Initiative, the landmark study of over 16,000 subjects that demonstrated the effects of conjugated estrogen plus a progestin on clinically important outcomes in postmenopausal women (4). The Women's Health Initiative debunked a number of myths about the miraculous effects that postmenopausal hormone therapy might have and led to much more judicious use of hormone therapy in women. In this millennium, testosterone has become one of the panacea snake oils for men who are older, frail, or just want to look better. Testosterone and other androgens are in the class of drugs known as “performance-enhancing,” a term that refers to athletic performance but resonates with the public mythology about the effects of testosterone in male performance in many aspects of life.

Despite the lessons learned about sex steroid hormone therapy in postmenopausal women from the Women's Health Initiative, we are threatened with a reprise of promiscuous prescription of sex steroid hormone therapy in aging men, obese men, diabetic men, and other groups of men with a high prevalence of low serum androgen levels. We are threatened with a mad “T” party. We could scarcely know less about the long-term effects of testosterone therapy on clinically meaningful outcomes in men. Contrary to Carrollian logic, we have proven with the absence of large, publicly funded trials that it is not “very easy to know more” about the diagnosis and management of androgen deficiency in men.

Into this rabbit hole, the authors of the 2010 Endocrine Society Guidelines on Testosterone Therapy in Men have boldly plunged. Their updates include a comprehensive review of the published literature, the addition of more than 40 recent references, and significant revisions of their recommendations regarding detecting prostatic cancer before initiation of androgen therapy and during maintenance therapy. They have subtly, but significantly, revised their recommendations for older men with low serum testosterone levels. The authors also have emphasized that long-acting opiates that are commonly prescribed may cause profound, symptomatic hypogonadism, and they have added a brief commentary on the controversy surrounding testosterone therapy for hypogonadal men with a history of prostate cancer that has been treated and deemed cured. Finally, the authors have included two new testosterone formulations that recently have become available (injectable testosterone undecanoate that may be injected at intervals up to 12–14 wk and a testosterone matrix patch that lasts 2 d after application).

There are some key differences in the new guidelines regarding detecting prostate cancer screening during androgen replacement therapy. The authors have appropriately specified that men younger than 40 yr do not need pretreatment screening for prostate cancer or monitoring during maintenance therapy. For men at least 40 yr old, the authors are vague about pretreatment screening for prostate cancer: “We suggest that clinicians assess prostate cancer risk in men being considered for testosterone therapy.” Although the authors do not explicitly recommend a pretreatment digital rectal exam plus measurement of serum prostate specific antigen (PSA), they imply that clinicians should use both to assess for baseline risk of prostate cancer. They recommend that clinicians estimate a patient's risk of prostate cancer by using a recently developed risk calculator (5) that includes digital rectal exam findings and a serum PSA level as well as family history, race, and prostate biopsy history.

This recommendation will be controversial. First, the risk calculator is not widely accepted because it is not clear how well it will perform in men in the general population (6). In addition, this risk calculator can only be used for men over age 55 (5,6). Secondly, screening for prostate cancer at any age is controversial and is particularly controversial for men ages 40 to 49. Although the American Urological Association recommends screening for prostate cancer starting at age 40, the American Cancer Society suggests that men at least 50 yr should be offered the opportunity for prostate cancer screening after careful discussion of the personal risks and benefits; men at higher risk of prostate cancer should begin this discussion at age 40 or 45 (7,8). The U.S. Preventive Services Task Force recently has recommended against prostate cancer screening in men 75 or older and has concluded that for men 75 yr old or younger there is “inadequate evidence to determine whether treatment for prostate cancer detected by screening improves health outcomes compared with treatment after clinical detection” (9).

Screening for prostate cancer is controversial because the benefit appears to be modest and the cost is significant. The results of two recently published prostate cancer trials (with over 200,000 patients combined) indicate that screening for prostate cancer may prevent one death due to prostate cancer per 1410 men screened and 48 treated, but prostate cancer screening does not reduce overall mortality (10,11). Screening for prostate cancer in hypogonadal men might be even less rewarding. Small studies suggest that exogenous testosterone therapy does not significantly increase prostate tissue testosterone levels or increase expression of genes associated with prostate cancer (12,13).

Most clinicians will decide to screen for prostate cancer before and during testosterone therapy in men over age 50 because testosterone may promote the growth of established invasive or metastatic prostatic cancer (14) and because of fear of litigation when a prostate cancer is detected incidentally some time after the initiation of testosterone therapy. The authors of the 2010 guidelines have highlighted an important and not widely known fact: serum PSA levels rise after testosterone therapy by an average of less than 0.5 ng/ml. Their recommendation to assess for possible prostate cancer for any serum PSA increase of at least 1.5 ng/ml in a year is conservative and is based on the wide natural variation in serum PSA levels in patients with benign prostatic hyperplasia. Even when the serum PSA level remains less than the upper limit of age-matched men, urological evaluation should be considered in any man whose serum PSA rises by at least 1.5 ng/ml in 1 yr.

To monitor for the development of prostate cancer, the guidelines authors continue to recommend a digital rectal examination in addition to serum PSA. This recommendation is not trivial because a digital rectal examination requires an office visit, is uncomfortable or embarrassing to many patients (and clinicians!), and is inexpertly performed by most clinicians. The evidence from randomized trials does not support the use of the digital rectal examination as a screening test (8). During testosterone therapy, screening with a serum PSA suffices.

In this 2010 update of the guidelines, the authors have made a subtle but important change to their recommendation regarding testosterone therapy for older men with symptoms suggesting hypogonadism plus low serum testosterone levels. The authors admit that there was considerable internal disagreement about the threshold of serum total testosterone level that would justify testosterone therapy in older men. Some experts on the panel favored treating only symptomatic men with very low total testosterone levels (e.g. <200 ng/dl), whereas others on the panel favored treating symptomatic men and total testosterone levels below the lower limit of normal. This debate pivots on whether there is a threshold serum testosterone level where the benefit exceeds the risk of testosterone therapy for symptomatic older men.

This disagreement is irresolvable based on current data, but it is refreshing to read guidelines that divulge disagreement among the panelists. Although guidelines are intended to be advisory based on an expert panel's collective assessment of the scientific literature, they are generally presented as monolithic consensus statements and often are perceived to define the standard of care. Guidelines seldom are as unanimous as they appear, and it is helpful for clinicians to read a candid disclosure about important disagreements. Knowledge about the controversy and the source of disagreement facilitates more informed and personalized care. For example, after reading these guidelines, a clinician who sees a 65-yr-old man with weakness, low libido, and a serum total testosterone of 280 ng/dl is more likely to acknowledge the uncertainty of benefit and risk; the clinician is more likely to engage the patient in a discussion that will inform the patient's choice.

The panel experts chose not to comment on the benefits and risks of testosterone therapy on cardiovascular events in men. Many, but not all, epidemiological studies suggest that men with lower serum testosterone levels have higher rates of cardiovascular events than men with higher serum testosterone levels (15,16,17,18,19), but there are no long-term large placebo-controlled studies of testosterone therapy that include cardiovascular outcomes. The authors cite a recent systematic review and a meta-analysis showing no differences in cardiovascular events and mortality in men (with baseline low serum testosterone levels) treated with testosterone vs. placebo, but these findings are based on short-term studies (20,21). Because the data are so scanty, it is appropriate that the panelists avoided making conclusions regarding the effects of testosterone therapy on cardiovascular risk.

There will be other points to debate in the guidelines: e.g. the role of free vs. total testosterone levels in the diagnosis of male hypogonadism, the value of determining bone densitometry at baseline and during androgen therapy, the threshold hematocrit to use as an exclusion criterion for testosterone therapy, and the relative benefit and risk of testosterone therapy in men with low testosterone levels plus diabetes mellitus. It is essential that we now begin large, long-term randomized trials of testosterone therapy to answer these questions and to measure clinically significant outcomes such as quality of life, functional capacity, prostate cancer, and cardiovascular events. These trials should address the major areas of controversy including whether older men, obese men, and other men with conditions commonly associated with low serum testosterone derive more benefit than harm from androgen supplementation therapy to normalize serum testosterone levels. In the meantime, we should commend the panelists for a careful review of the scientific literature and making clear, reasonable recommendations that will help practicing clinicians. Most importantly, they framed each recommendation as a good-faith estimation of the benefits and risks for the individual patient. We asked for their opinions, and they made the recommendations appropriately personal—focused on the individual and not broad sweeping generalizations.

“At any rate I'll never go there again!” said Alice as she picked her way through the wood. “It is the stupidest tea party I ever was at in all my life!” Once more she found herself in the long hall, close to the little glass table. “Now I'll manage better this time.”

With clear, well-reasoned guidelines such as these, we'll manage better, too.